Creating a Satellite Composite Basemap

This example utilizes the land/sea bitmask provided as a grid function within GEMPAK grid diagnostics. The resolution of the bitmask is 0.5 degree, and is adequate for the resolution of the N. Hemispheric Composite satellite product.

Tools Used

Gempak Programs

• IMG2GD
• GD2IMG

Data Used

• IR Satellite Imagery

Steps

1. Convert current IR satellite image to grid using IMG2GD

   
       PROJ     = STR/90;-105;0
       GRDAREA  = -21;-150;-21;30
       KXKY     = 800;800
       GRDNAM   = irsat
       GDFILE   = nhem_sat.gem
       GLEVEL   = 0
       GVCORD   = none
       GDATTIM  =
       CPYFIL   =
       MAXGRD   = 200
       IMGTIM   = last
       CALIMG   = no
       IMGFIL   = NHEMMULTI_IR
       GEMPAK-IMG2GD>r
       satellite image date/time 070530/1200
      

2. Create a composite image satellite and land/sea mask using GD2IMG
   • Use IRSAT values where brightness value is greater than 112 (eg 113 to 255, or colder than 0C). Subtract 112 from pixel values to produce values from 1 to 143 and set remaining values to 1.
   • Create a 0/1 boolean grid of land/sea.
   • Multiply the boolean 1/0 grid by 143 and add to the transformed IRSAT values as above.
   • The result is a grid of values ranging from 1 to 143 over land and 144 to 286 over the ocean.
   • Store the image as a gini image (Polar Stereographic projection 800 x 800).
   • Use a calibration range for band 10 for colors 0 to 94 representing pixel values 0 to 286.

         GDATTIM  = last
         GDFILE   = nhem_sat.gem
         GLEVEL   = 0
         GVCORD   = none
         SCALE    = 0
         GFUNC    = add(mul(bool(sea),143),miss(sub(sle(sgt(irsat,112),255),112),1))
         PROJ     = STR/90;-105;0
         GRDAREA  = -21;-150;-21;30
         KXKY     = 800;800
         CPYFIL   =
         SATFIL   = COMP_YYYYMMDD_HHNN
         CALINFO  = 99/10/SAT,0,94,0,286
         WMOHDR   = TICZ99/CHIZ/
         GEMPAK-GD2IMG>r
        

3. Create a color lookup table representing land, ocean, and shades of IRSAT clouds for the 2 ranges of data above. The color values 0 to 94 will be divided into 2 groups of 47 with the last color entry representing missing data. Each section of 47 colors will start with the color for land or ocean and end with the maximum value of white (RGB 255,255,255). The remaining colors will be evenly ramped from the starting color to the ending color (a short program saved time for this step). You can download the color table upc_t_s.tbl generated for this composite satellite image.
4. Add an image type entry for this band 2**(10 - 1) image for sensor 11 (Gini ID 99 above) into GEMTBL/sat/imgtyp.tbl if it does not already exist.

      IRSAT-BASE           COMPOS        0     94     11   2**9      1 upc_t_s.tbl
   

5. Display the composite image in GEMPAK programs as a satellite image and overlay data as desired.

   
       MAP      = 1
       GAREA    = dset
       PROJ     = sat
       SATFIL   = COMP_20070530_1200
       LUTFIL   = upc_t_s.tbl
   
      

   

6. To create a realtime product, schedule the conversion of the IR image to grid and generation of the composite image.