This is a series of four images of the K fragment in the 7.7 micron
methane region at a resolution of 9000 using KEGS on the Kuiper Airborne 
Observatory.  In each image, wavelength increases from 7.685 to
7.722 microns from left to right.  The y axis represents a long-slit
spatial cut through Jupiter.  From top to bottom, these images were 
taken at UT= 10:28, 10:39, 10:53, and 11:04.  The fireball is clearly
evident in the second image; the methane emission increased by about
a factor of 25 compared to the pre-impact emission (top image).  

The spatial scale is 4"/pixel. Our point spread function is ~7" due to 
motion of the airborne telescope. The brightness contours range from 1
to 43 arbitrary units in steps of 2.

The 7.7 micron region is a sensitive thermometer of the jovian stratosphere
sounding levels between 1 microbar and 10 millibars, far above the visible
clouds. This region is totally blocked by methane in the Earth's atmosphere -
it cannot be observed from ground-based telescopes. The KAO is the only
observatory capable of investigating this spectral region.

The high spectral resolution of the KEGS instrument reveals individual
emission lines of C-12 and C-13 CH4. At least 2 unknown emission lines
were observed in the spectrum of the fireball. These may be due to high
excitation states of CH4 or due to a different species altogether.

An important objective of our KAO investigation is to determine the level
at which the cometary fragments exploded. We had 2 water channels (at 22.6
and 23.9 microns) in addition to the temperature channel at 7.7 microns.
We alternated between these 3 grating positions throughout the flight. The
7.7 micron results are presented here because of the dramatic nature of the
data. The 2 water channels are more difficult to interpret. The continuum is
produced by pressure-induced H2 opacity which sounds the temperature of the
upper troposphere (~200 millibars). The continuum was observed to change by
a small amount, but this may be entirely due to changes in viewing geometry
during the observations. There was no dramatic change in either the continuum
level or in the depth of the water lines. Subtle changes will have to be
investigated carefully. Our extremely preliminary examination of the water
data suggest the following: the explosion did not take place at the 10 bar
level on Jupiter. Otherwise we would have seen strong H2O emission lines in
each of our 2 water channels due to the transport of water vapor in a rising
plume. Second, the bulk of the energy was not deposited near the 200-millibar
level; otherwise, the continuum level at 22.6 and 23.9 microns would have
increased dramatically. 

Our positive detection of enhanced CH4 emission and the non-detection of
jovian water and  non-detection of elevated temperatures at 200 millibars
suggest that the explosion of the G and K fragments took place in the
stratosphere. We emphasize that these results are very preliminary as we
have looked at only a small fraction of our spectra in a hangar at the
airport in Melbourne, Australia after an exhausting series of flights.

We would like to thank everyone associated with the KAO as well as our
hosts in Australia for a successful flight program.

Gordon Bjoraker (NASA/GSFC)
Terry Herter, Susan Stolovy, George Gull, and Bruce Pirger (Cornell Univ)