Megan, I looked at your attached images, and the 2nd image which I believe is supposed to be your satellite image while showing the color bar, appears to have the wrong LUTFIL defined. Ensure you either have set LUTFIL=default to use the color lookup table define in satimg.tbl, or you have defined LUTFIL=ir_upc2.tbl in GDPLOT2). Otherwise, I do not see the number of colors in your plot that I do here. Check your nex2gini.tbl for any typo as well- the two sections of the curve are separated by the ";" character. I have attached a GIF image that I produced here showing the satellite image I created using your CTMPC grid and the modified Celsius nex2gini.tbl entries of: CTMPC 16 2 C,255,176,-110,-31;175,0,-30.5,57 TICZ99 TMPC I have overlaid the image with the TYPE=p plot of data values from your grid using IJSKIP=5 for readability. I do see gray areas to the northwest of the storm with values from 18 to 29C. I have also attached the GINI image I used, so you should be able to duplicate my image if you had a typo in your nex2gini.tbl. If the problem was in your LUTFIL setting in the display, then you would probably see a similar issue with not matching the display of the GIF that I sent. Steve Chiswell Unidata User Support Where previously I had sent you the TMPK values. > Dr. Chiswell, > > I have a question about the mapping of temperature to pixel at the higher > temp values. I've noticed that there's a lack of the the warmer/light > gray colors in my satellite images. > I've included some figures as an example of what I mean. The observed IR > has a lot of the lighter grays (5 to 20 C) around the storm. My IR image > shows a pretty abrupt change from the colder temps (below 0) to a black > background. > I checked output from the algorithm before it is converted to image and > just plotted my homemade satellite IR table with some contours; it's the > gempak.gif file. The warmest temperatures it shows are between 20 to 25 > outside the northwest quadrant of the storm and within the eye (the sign's > flipped on the contours so that the color table would go up for decreasing > temperature). So shouldn't I have a light gray color in those areas, > instead of the black that shows up when it's mapped to a pixel? > I'm not sure I understand the IR temperature curve that is in > nex2gini.tbl. Is this just because the IR temperature is plotted on a > logorithmic scale for the colder temps and a linear scale for the warm > values? > > Thanks for all your help, > Megan > > >> Dr. Chiswell, > >> > >> I have a question of gd2img. The satellite observations that I'm using > >> to > >> compare with my model-simulated satellite are all in celsius. I have > >> both > >> celsius and kelvin versions of cloud top temperatures in my gempak files > >> before they're converted to gini with gd2img. Can I also use gd2img to > >> convert the celsius cloud top temperatures? > > > > Yes, you can use any variable for GFUNC as long as you create a > > nex2gini.tbl > > entry such as: > > CTMPC 15 2 K,255,176,-110,-31;175,0,-31.5,15 TICZ99 TMPK > > > > The band number of n=15 is what is used in imgtyp.tbl column 6 > > as 2**n-1 (so that would be 2**14), where I had previously > > sent you 2**15 for band=16 (or you can just redefine CTMPK to CTMPC > > using the same band number). > > > > Steve Chiswell > > Unidata User Support > > > >> I attempted to put a table entry for a celsius variable (CTMPC) in > >> nex2gini.tbl, by subtracting 273 off all the values for the kelvin cloud > >> top temps. Is that correct? I'm also not sure what values to put into > >> imgtyp.tbl for CTMPC. > >> Thanks! > >> Megan > >> > >> > Megan, > >> > > >> > I put together a completely standalone tarfile (attached to this > >> message) > >> > that should run without any upgrade of your previous versions. > >> > > >> > First thing is to see if your Redhat can run the binary I built here > >> under > >> > fedora core 5. > >> > > >> > Unpack the tarfile (it will create its own "ncsu" subdirectory where > >> you > >> > unpack it. > >> > In the directory are the nex2gini.tbl file that defines the mapping > >> from > >> > temperature to pixel value: > >> > CTMPK 16 2 K,255,176,163,242;175,0,242.5,330 TICZ99 > >> TMPK > >> > > >> > The 2 part IR mapping is pixel values 255 -> 176 convert to 163K -> > >> 242K, > >> > 175 -> 0 convert to 242.5K -> > >> 330K > >> > These match the standard IR curve used for10.7 micron imagery in > >> > $GEMPAK/source/gemlib/im/imbtot.f > >> > and GEMPAK/source/gemlib/im/imttob.f. > >> > > >> > The CTMPK in this entry must match the name of your grid function! > >> Band > >> > used will be "16" which is used in the $GEMTBL/imgtyp.tbl entry I sent > >> > you previously as 2**(band-1) (aka 2**15). > >> > > >> > You will be able to run the program within the ncsu directory since > >> all > >> > tables are searched > >> > in the current working directory. Otherwise, the nex2gini.tbl would be > >> > found in > >> > $GEMTBL/unidata/nex2gini.tbl, the gd2img.pdf would be found in > >> > $GEMPDF/gd2img.pdf. > >> > > >> > The program can be run as: > >> >> gd2img > >> > GDATTIM Grid date/time f024 > >> > GDFILE Grid file satir_f24.gem > >> > GLEVEL level 0 > >> > GVCORD Vertical coordinate none > >> > GFUNC Gridded parameter CTMPK > >> > SCALE Output scaling 0 > >> > PROJ Map projection/angles/margins|dr > >> > GRDAREA Area covered by grid > >> > KXKY Number of grid points in x;y > >> > CPYFIL Grid file whose navigation is to satir_f24.gem > >> > SATFIL output file name testimg.gini > >> > > >> > Since your grid is a 4x4km LCC grid, I used that in cpyfil for > >> > the output image projection. Otherwise, a GRDAREA, PROJ, and KXKY > >> could > >> > be specified but it has to be a lambert tangent cone with dx=dy, but > >> > you probably don't want to have to figure that out unless you change > >> the > >> > model resoltion! > >> > > >> > The GFUNC name CTMPK must match the entry in nex2gini.tbl > >> > > >> > Go ahead and give this a try. As I mentioned, its under development so > >> it > >> > spits out some debug information > >> > and the help file is basically non-existant, but improving 100% all > >> the > >> > time ;-) > >> > > >> > Let me know if you can get the program to run, and I'll keep working > >> on > >> > cleaning up > >> > the interface and input. Of course, you will eventually want to run a > >> > production version in an > >> > up to date GEMPAK release in the future. > >> > > >> > Steve Chiswell > >> > Unidata User Support > >> > > >> > > >> > > >> > > >> >> Dr. Chiswell, > >> >> We are running Red Hat Enterprise Linux, 4 or 5. I am currently > >> using > >> >> gempak v5.6. If that's too old, I think 5.8.3 is also compiled on > >> the > >> >> cluster I'm using, so I could use it instead. > >> >> Thanks again for your help, > >> >> Megan > >> >> > >> >> >> Mr. Chiswell, > >> >> >> I took a look at that test image you sent, and that's exactly what > >> I > >> >> >> wanted. > >> >> >> I'd like to generate a lot more of these images from my model > >> output. > >> >> I > >> >> >> was wondering if you could send me the code you used. Could I > >> >> install > >> >> >> gd2img by itself or would I need a whole new gempak version? > >> >> >> Thanks, > >> >> >> Megan > >> >> > > >> >> > Megan, > >> >> > > >> >> > I have the gd2img directory in the source distribution of 5.10.1, > >> but > >> >> > didn't > >> >> > configure it to be built in the binary distributions since it is > >> still > >> >> > under development. > >> >> > I have been testing it out to create topo images from dem data, but > >> >> > created the > >> >> > TMPK to pixel value entry in the table for your data. I will make > >> it > >> >> part > >> >> > of the standard > >> >> > build for 5.10.2; I just need to add some user friendliness to the > >> >> > program. > >> >> > > >> >> > I can send you the program to run without upgrading your > >> distribution > >> >> if > >> >> > you can tell me what OS etc you are running. Do you know what > >> version > >> >> of > >> >> > GEMPAK you are > >> >> > currently running? > >> >> > > >> >> > Dr. Steve Chiswell > >> >> > NC State 87, 90, 94 > >> >> > Unidata User Support > >> >> > > >> >> > > >> >> > > >> >> >> > >> >> >> > >> >> >> > Megan, > >> >> >> > > >> >> >> > I created the attached GINI format satellite image from your > >> data. > >> >> >> > To display the image, you will need to add the following line to > >> >> your > >> >> >> > $GEMTBL/sat/imgtyp.tbl file (or imgtyp.tbl file in your working > >> >> >> > directory): > >> >> >> > > >> >> >> > CTMPK IR_TMPK 0 255 11 2**15 1 > >> >> >> > ir_upc3.tbl > >> >> >> > > >> >> >> > Note use only spaces in that file, not tabs. > >> >> >> > > >> >> >> > The conversion process is somewhat manual for this process since > >> I > >> >> >> have to > >> >> >> > generate a > >> >> >> > mapping of pixel values to gridded data values. This is > >> something I > >> >> >> need > >> >> >> > to add for > >> >> >> > users in a working program, but doing this with your data helps > >> me > >> >> get > >> >> >> to > >> >> >> > that > >> >> >> > point - I did discuss this type of project with Gary Lackmann > >> here > >> >> >> last > >> >> >> > summer, > >> >> >> > so its a work in progress. > >> >> >> > > >> >> >> > I need to find out from you if this mapping of brightness > >> >> temperature > >> >> >> to > >> >> >> > pixel > >> >> >> > is what you are looking for (its the standard 2 part temperature > >> >> curve > >> >> >> > used for > >> >> >> > IR satellites), and if not, I'll need to get this worked out to > >> >> >> provide > >> >> >> > you with the > >> >> >> > program. > >> >> >> > > >> >> >> > I displayed this with your grid points and contours in gdplot2 > >> and > >> >> the > >> >> >> > image lines up > >> >> >> > with the points, so my first pass seems to be in the right > >> >> direction. > >> >> >> > > >> >> >> > Let me know how this works for you..... > >> >> >> > > >> >> >> > Steve Chiswell > >> >> >> > Unidata User Support > >> >> >> > > >> >> >> > > >> >> >> >> Here's the grid navigation from gdinfo... > >> >> >> >> GDFILE Grid file satir_f24.gem > >> >> >> >> LSTALL Full list flag YES > >> >> >> >> OUTPUT Output device/filename T > >> >> >> >> GDATTIM Grid date/time last > >> >> >> >> GLEVEL Grid level 0 > >> >> >> >> GVCORD Grid vertical coordinate none > >> >> >> >> GFUNC Scalar grid CTMPK > >> >> >> >> Parameters requested: > >> >> >> GDFILE,LSTALL,OUTPUT,GDATTIM,GLEVEL,GVCORD,GFUNC. > >> >> >> >> GEMPAK-GDINFO>r > >> >> >> >> > >> >> >> >> GRID FILE: satir_f24.gem > >> >> >> >> > >> >> >> >> GRID NAVIGATION: > >> >> >> >> PROJECTION: LCC > >> >> >> >> ANGLES: 27.0 -85.3 27.0 > >> >> >> >> GRID SIZE: 159 159 > >> >> >> >> LL CORNER: 15.83 -82.64 > >> >> >> >> UR CORNER: 21.20 -76.45 > >> >> >> >> > >> >> >> >> GRID ANALYSIS BLOCK: > >> >> >> >> ANALYSIS TYPE: BARNES > >> >> >> >> DELTAN: 0.073 > >> >> >> >> DELTAX: -9999.000 > >> >> >> >> DELTAY: -9999.000 > >> >> >> >> GRID AREA: 15.00 -84.00 22.00 -75.00 > >> >> >> >> EXTEND AREA: 15.00 -84.00 23.00 -75.00 > >> >> >> >> DATA AREA: 15.00 -84.00 23.00 -75.00 > >> >> >> >> > >> >> >> >> Also, the file (satir_f24.gem) is posted at.. > >> >> >> >> http://tempest.meas.ncsu.edu/megan > >> >> >> >> if you'd like to have the actual file. > >> >> >> >> > >> >> >> >> Basically, I'm working on generating a model-simulated > >> satellite > >> >> IR > >> >> >> >> product, kind of like model-simulated radar. But I'd like to > >> use > >> >> the > >> >> >> >> satellite color tables to make an observational comparison. > >> I'd > >> >> >> >> eventually like to try and use the model-simulated IR on grids > >> of > >> >> >> >> different navigation, but I started out with this grid because > >> >> it's > >> >> >> 4-km > >> >> >> >> grid spacing, similar to satellite resolution. > >> >> >> >> > >> >> >> >> Thanks for your help! If there's any other information you'd > >> like, > >> >> >> just > >> >> >> >> let me know. > >> >> >> >> > >> >> >> >> Megan > >> >> >> >> > >> >> >> >> > >> >> >> >> > Megan, > >> >> >> >> > I have been working on a gd2img program in the source > >> >> distribution > >> >> >> >> which > >> >> >> >> > can convert a grid to an image using a specified satellite id > >> >> and > >> >> >> >> > channel number with a projection remapped into one that the > >> >> image > >> >> >> >> format > >> >> >> >> > can handle. > >> >> >> >> > > >> >> >> >> > Can you send me some information about what grid > >> >> >> projection/navigation > >> >> >> >> > you have so I can see if this will be sufficient for your > >> >> >> application? > >> >> >> >> > The GINI format is somewhat limited specifying a projection > >> and > >> >> >> pixel > >> >> >> >> > shape, so might have to use an AREA format instead. Please > >> send > >> >> >> this > >> >> >> >> to > >> >> >> >> > address@hidden. > >> >> >> >> > > >> >> >> >> > Steve Chiswell > >> >> >> >> > Unidata User Support > >> >> >> >> > > >> >> >> >> > On Fri, 2007-03-30 at 10:25 -0400, Megan Gentry wrote: > >> >> >> >> >> I have a GEMPAK grid file that I'd like to convert to image > >> >> format > >> >> >> >> and > >> >> >> >> >> view using the satellite IR color tables. I was wondering > >> what > >> >> >> the > >> >> >> >> best > >> >> >> >> >> way is to convert a grid file to an image file that can be > >> >> viewed > >> >> >> >> using > >> >> >> >> >> the IR color tables? > >> >> >> >> >> Thanks, > >> >> >> >> >> Megan > >> >> >> >> >> > >> >> >> >> >> ---------------------------------------------------- > >> >> >> >> >> Megan S. Gentry > >> >> >> >> >> Graduate Research Assistant, Forecasting Lab > >> >> >> >> >> Department of Marine, Earth and Atmospheric Sciences > >> >> >> >> >> North Carolina State University > >> >> >> >> >> > >> >> >> >> >> =============================================================================== > >> >> >> >> >> To unsubscribe gembud, visit: > >> >> >> >> >> http://www.unidata.ucar.edu/mailing-list-delete-form.html > >> >> >> >> >> =============================================================================== > >> >> >> >> > -- > >> >> >> >> > Steve Chiswell <address@hidden> > >> >> >> >> > Unidata > >> >> >> >> > > >> >> >> >> > >> >> >> >> > >> >> >> >> ---------------------------------------------------- > >> >> >> >> Megan S. Gentry > >> >> >> >> Graduate Research Assistant, Forecasting Lab > >> >> >> >> Department of Marine, Earth and Atmospheric Sciences > >> >> >> >> North Carolina State University > >> >> >> >> > >> >> >> >> > >> >> >> > > >> >> >> > > >> >> >> > Ticket Details > >> >> >> > =================== > >> >> >> > Ticket ID: FWQ-357482 > >> >> >> > Department: Support > >> >> >> > Priority: Normal > >> >> >> > Status: Closed > >> >> >> > >> >> >> > >> >> >> ---------------------------------------------------- > >> >> >> Megan S. Gentry > >> >> >> Graduate Research Assistant, Forecasting Lab > >> >> >> Department of Marine, Earth and Atmospheric Sciences > >> >> >> North Carolina State University > >> >> >> > >> >> >> > >> >> > > >> >> > > >> >> > Ticket Details > >> >> > =================== > >> >> > Ticket ID: FWQ-357482 > >> >> > Department: Support GEMPAK > >> >> > Priority: Normal > >> >> > Status: Closed > >> >> > > >> >> > > >> >> > >> >> > >> >> ---------------------------------------------------- > >> >> Megan S. Gentry > >> >> Graduate Research Assistant, Forecasting Lab > >> >> Department of Marine, Earth and Atmospheric Sciences > >> >> North Carolina State University > >> >> > >> >> > >> > > >> > > >> > Ticket Details > >> > =================== > >> > Ticket ID: FWQ-357482 > >> > Department: Support GEMPAK > >> > Priority: Normal > >> > Status: Closed > >> > >> > >> ---------------------------------------------------- > >> Megan S. Gentry > >> Graduate Research Assistant, Forecasting Lab > >> Department of Marine, Earth and Atmospheric Sciences > >> North Carolina State University > >> > >> > > > > > > Ticket Details > > =================== > > Ticket ID: FWQ-357482 > > Department: Support GEMPAK > > Priority: Normal > > Status: Closed > > > > > > > ---------------------------------------------------- > Megan S. Gentry > Graduate Research Assistant, Forecasting Lab > Department of Marine, Earth and Atmospheric Sciences > North Carolina State University > Ticket Details =================== Ticket ID: FWQ-357482 Department: Support GEMPAK Priority: Normal Status: Closed
Attachment:
upc_display.gif
Description: GIF image
Attachment:
testimg.gini
Description: Binary data
