The outline of the major ice shelves in Antarctica is missing from the GMT datasets. I have included them and made them available in add.tar.gz and greenland.dat.gz.

For example, here is figure showing the Antarctic and Greenland ice sheets (the GMT bash script to generate this figure is 2poles.gmt here.

# ... some set up not shown

# draw the world 
pscoast -R0/360/-90/90 -JG-75/41/4.5i -B45/30 -Dc -A10000 \
    -G100/100/100 -X7c -Y10c -K -O >>$out
gunzip -c greenland.dat.gz | \
    psxy  -JG -R -M -G0/100/240 -K -O >>$out
    GMT uses a terse set of command line flags. In this case, we have:
  • -R0/360/-90/90, which sets the region to the whole globe
  • -JG-75/41/4.5i which sets the projection to Azimuthal Orthographic with a projection center of 75W, 41N (approximately in Pennsylvania)and the width of the map is 4.5 inches.
  • -B45/30 sets the drawing of gridlines to every 45 degrees in longitude, and every 30 degrees in latitude
  • -Dc says to only draw coarse features
  • -A10000 says to limit water features to those larger than 10,000 km square
  • -G specifies the color as gray
  • and the -X and -Y flags offset the map (units are 'c'=cm and 'i'=inches). In GMT, the origin is at the bottom left of the page, and the positive x- and y-axes are to the right and upwards, respectively.

The 2nd line uncompresses greenland.dat.gz and draws that outline in blue (using the same -R and -G parameters as before---if you give bare options, GMT will use the the most recent settings...).

#... draw the southern hemisphere and Antarctica in gray
pscoast -R0/360/-90/90 -JG-75/-41/4.5i -B45/30 -Dc \ 
    -A10000 -G100/100/100 -X-5c -Y-8c -K -O >>$out
# ... and add my outline on top in blue
gunzip -c ADDant.txt.gz | psxy -JG -R -M \
  -G0/100/240 -K -O >> $out

The GMT default Antarctic outline (drawn by the first command) includes the ice shelves. Here I use a projection center with the same longitude, but with a latitude of -41S (in the eastern Pacific off the coast of S. America...).

The 2nd line uncompresses my outline (which excludes the Ross, Filchner-Ronne, Amery, and Larsen ice shelves) and draws that over the first outline.

Note that the offset of this map is relative to the previous drawing location (which was, if you recall, 7c/10c from the origin at the bottom left). Thus the negative offsets moves the 2nd globe down and to the left.