CASPIR observations of the collision of Comet Shoemaker-Levy 9 with Jupiter

The CASPIR near-infrared camera was used to record the impacts of fragments C, D, G, K, N, R, V, and W of Comet Shoemaker-Levy 9 with Jupiter. We present and discuss 2.34 item lightcurves and 1-5 mu m images for six of these impacts. Leader emission was seen up to 3.5 min prior to the G and K impacts which we interpret as due to the early impact of cometary coma material that extended along the fragment orbit prior to impact. First precursors were seen for the G and K impacts immediately prior to impact, due to thermal emission from the inbound meteor or its trail. Second precursors were seen similar to 1 min after impact for C, D, G, K, and W and lasted for <90 sec, This thermal emission was seen as the rising hot plume first came into direct view from Earth. The emission then faded for the smaller impacts, but slowly increased in strength for the larger G and K impacts in what we identify as a third precursor, We propose that this may be evidence for continuous ejection of hot material. Two classes of lightcurves are identified based on the presence of this third precursor emission for the larger G and K impacts and its absence in the lightcurves for smaller fragments. The main brightening began similar to 6 min after impact for all six of the fragments studied, independent of the location of the impact site behind the limb. We interpret this as the minimum ejecta flight time before fall back on the atmosphere. A simple ballistic model shows that material ejected toward the limb landed within our direct view at that time, Shoulders seen in several lightcurves as the main peak declined are possibly due to material that bounced on first re-entry, Large rings were seen around the G, K, and possibly C impact sites in 3-4 mu m images. The G and K rings expanded at a speed of similar to 1.4 km sec(-1) until they reached radii of at least 18,000 km, These rings lie outside the debris crescents seen in the optical by LIST and were not seen at shorter wavelengths, The origin of these large rings remains uncertain, although they may be due to ejected material sliding over the top of the atmosphere. The emission may be from an organic species that must emit or scatter 3-4 mu m photons efficiently while remaining dark at other wavelengths, Transient northern amoral emission was seen in 3.30-3.99 mu m images immediately following the G, K, R, and W impacts, The spectral distribution of this emission is broadly consistent with H-3(+) emission. (C) 1996 Academic Press, Inc.