Skip Navigation: Avoid going through Home page links and jump straight to content

Abstracts from IAU Colloquium 156


Here are abstracts for all of the invited talks at the IAU Colloquium 156
in Baltimore, Maryland, during May 9-12, 1995.
They are listed in the order
in which they will be presented.

Speaker: Torrence Johnson (
Title: The first minutes: Galileo's SL9 Observations

Speaker:Heidi Hammel (
Title: Observations during the First Two Hours after Impact - Plumes
        and Fresh Sites
During the first two hours after impact, numerous phenomena were 
observed with telescopes on Earth, in orbit, and in space. The 
primary events were: impacts themselves, rise and fall of large 
plumes of ejected material, and atmospheric waves; also of interest
is the characteristic morphology of fresh impact sites. HST recorded
initial impact phenomena for fragments A, G, and W (the E impact 
occurred just prior to the HST observation window), based on
comparison of event timing with data from Galileo instruments and 
ground-based observations.  For those three events and for the E 
impact, plumes were detected rising above the limb of Jupiter.
The plumes reached heights of more than 3000 kilometers within 
5 minutes, independent of the size of the resulting impact site.
Several ground-based detections of plumes have also been
reported.  Some HST plumes were visible in Jupiter's shadow; 
above the shadow, they reflected sunlight. Plume development and 
collapse correlates with strong infrared emission near the jovian 
limb, leading to the interpretation that the IR brightness was 
created by the fall-back of plume material from high altitude.
The initial site morphology was remarkably similar for all
medium-to- large impacts: a dark streak and a larger 
crescent-shaped ejecta to the southeast.  These crescents appear 
to be thin ejecta blankets, and sometimes extended up to 13,000 km 
from the impact sites.  This morphology was seen not only by HST 
but also in ground-based images.  Small impact sites typically 
only showed a dark patch (no ejecta).  For medium-to-large impact 
sites that were imaged by HST within 2.5 hours of impact 
(A, E, G, Q1, and R), transient expanding rings were detected, 
most likely caused by horizontal propagation of atmospheric
waves.  Multiple rings were seen for impacts G and E. 
The current interpretation of these observations will be discussed.

Speaker: Phil Nicholson (
Title: Ground-based Observations of Impact Phenomena
Observations of the impacts of fragments of SL9 with Jupiter were made
from an extensive network of ground-based telescopes in Europe, South
Africa, North and South America, Hawaii, Japan, Australia, and the
Antarctic, at wavelengths ranging from the visible to decametric. In
this review, we will concentrate on visible to mid-infrared
observations of the prompt impact phenomena, limited to the first 60
min after the individual impacts.  The most extensive sets of data
were obtained in the near-infrared, especially at 2.3 and 3.5 microns,
utilizing strong methane absorption bands in Jupiter's spectrum to
enhance the contrast of the impact features.  By a happy coincidence,
this was also near the peak of thermal emission from the impact plumes
themselves and from their hot remnants, seen as the sites rotated onto
the planet's visible disk 5--10 min after each impact. To date,
successful earth-based observations have been reported for 16 impacts:
A, B (Keck only), C, D, E, G, H, K, L, N (Siding Spring only), Q2, Q1,
R, S, V (Palomar only), and W.  The predicted impact times for
fragments F, J, M, P2, T and U passed without convincing evidence of
any impacts, and no corresponding post-impact features were visible in
HST images for these fragments. In almost all cases, the maximum
near-IR signal (the bright `flare' seen in many published images)
occurred 10 -- 15 min after the actual impact time (as inferred from
direct observations by Galileo or plume imaging by HST), and is most
plausibly attributed to thermal emission from cometary and jovian
material shock-heated to temperatures of order 500 -- 1000 K by the
re-entry of the impact plume into the atmosphere. In several instances
(notably events A, G, H, K, L, Q1, R and W), one or more precursor
flashes was observed prior to the main flare and within a minute or
less of the inferred impact time. In particular, impacts H, L, K and R
showed two distinct precursor flashes, the second and brighter being
almost certainly due to the hot plume first appearing above the
planet's limb. The first flash was seen in each case 20 -- 30 sec {\it
prior to} the initial detection by Galileo, and has been interpreted
as the bolide entering the planet's upper atmosphere, although
reflection of light from the explosion itself remains a (remote)
possibility.  Another feature common to several of the near-IR and
mid-IR light curves (H, K, L, Q1, R) is a secondary maximum which
occurs $\sim10$ min after the bright flare's main peak. This
unexpected `bounce' may be due to an oscillation in Jupiter's
stratosphere triggered by the impact, or even to a literal rebound of
the re-entering ejecta.  Additional observations include a possible
reflection of the E impact flash from Europa, and the direct visual
imaging of the H plume above the limb.

Speaker: Jacques Crovisier (
Title: The Composition and Structure of Comet D/Shoemaker-Levy 9
With Shoemaker-Levy 9 we were the lucky witnesses of one infrequent
event: the evolution of the fragments of a disrupted comet, and of a unique
event in the history of humanity: the high-velocity impacts of these massive
fragments in the atmosphere of a giant planet. Unusual physical and chemical
processes were observed. What did we learn on the physical and chemical nature
of the comet from these events ? 

I will try to answer this question on the basis of the information
obtained from the spectroscopic and imaging observations of the comet prior
to the impact and from the observations - mainly spectroscpic - of the 
impact sites.

It seems that the wealth of information recorded could not give us
direct clues to the structure (mass, density, size) of the comet or to its
chemical composition. The definitive conclusions will only come from the 
cooperation of observers and modellers trying to put all the pieces of the
puzzle together. 

Speaker: David Crawford (
Title: Models of Fragment Penetration and Fireball Evolution
Numerical simulations of the impact of Comet Shoemaker-Levy 9 on
Jupiter, if calibrated against the observations, can potentially help
determinine such things as comet fragment mass, density, size and
composition. Prior to the event, numerous researchers performed
computational studies of the impact events and produced a wide array of
predictions. Since the event, some of the modeling results are
converging, yet some fundamental differences remain. In this review, I
will attempt to describe modeling efforts from groups at CalTech, Los
Alamos and Sandia that describe fragment penetration and fireball
evolution. In particular, I will attempt to determine why different
groups appear to get different penetration depths with similar fragment
starting models. At Sandia, we have used the CTH Eulerian shock physics
code to study 2-D and 3-D representations of the events. From this
modeling, we have reached several conclusions: 1) Growth of the
fireball during the first few minutes primarily depends on energy
deposited at relatively high altitude in the Jovian troposphere and
stratosphere. Energy deposited below this level (by large, dense
impactors, for example) has relatively little influence on early-time
fireball growth. 2) At high altitude in the Jovian atmosphere, energy
deposition is primarily a function of the cross-sectional area of the
impactor; therefore, energy deposited by a 3-km diameter cloud of icy
debris is about the same as that due to a 3-km sphere of solid ice. 3)
Based on observed plume heights and trajectories, we estimate that the
largest fragments of S-L 9 had effective diameters between 1 and 3 km
in diameter at the time of impact. Under this scenario, uniform plume
heights imply consistent diameters, but not necessarily equal masses.
This work funded by NSF under agreement No. 9322118 and performed at
Sandia National Laboratories supported by the U.S. Dept. of Energy
under contract DE-AC04-94AL85000.

Speaker: Mordecai-Mark Mac Low (
Title: Entry Models vs. Observations : What Have We Learned?
The size and density of impacting fragments of Comet Shoemaker-Levy 9
determined the depth at which they explosively released their kinetic 
energy to the atmosphere.  The size and depth of these explosions 
determines a number of observables.  We describe a simple analytic model 
of the entry that can predict the explosion depth and its verification by 
numerical computations using an astrophysical hydrodynamics code, ZEUS.  
The analytic model uses a quasistatic treatment of the spreading of the
impactor to find its energy deposition.  Our high-resolution computations 
(10-100 m resolution on a 10 x 100 km grid) show that the analytic and 
numerical models agree on the predicted point of maximum energy release, 
although some cometary material penetrates deeper into the atmosphere as 
suggested by Crawford et al.  Our analytic model succeeds because the 
timescale predicted is the same as the timescale for fragmentation by 
Rayleigh-Taylor instabilities, as shown by Svetsov et al.  An attempt to 
determine the size of the impactors was made by Asphaug & Benz by comparing 
the predictions of tidal breakup models with observations of the comet train. 
We find that objects of the size they found, having a few x 10^27 ergs of 
kinetic energy (eg sizes of 0.5 - 1 km at densities of 0.5 - 0.7 g/cm^3) can 
coherently explain at least six major observational results.  These
include the faintness of the fireballs in visual light, the infrared 
light curves, the abundance of diatomic sulfur in the plumes, the lack of 
atmospheric water in the plumes, the lack of detectable seismic waves, and 
the morphology of the spots.

Speaker: Kevin Zahnle (
Title: Models of Fallback
The impact of a typical Shoemaker-Levy 9 fragment produced 
three light peaks as seen from Earth. The first peak is 
related to the entry of the fragment into the jovian
atmosphere. The second peak occurs when the exploding 
fireball rises above Jupiter's limb into direct view from
Earth. The third peak, much the brightest, occurs when the
ejecta plume falls back on the atmosphere. Here we present
a simple, highly idealized model of a ballistic plume,
which we then use to fit the observed light curve of the 
R impact as recorded at Mauna Kea and Mt. Palomar. 
The nominal R fragment has diameter 500 m and
mass 3x10^13 g, with energy release of 5x10^26 ergs.
The largest events were about three times more energetic.
These sizes agree with those deduced from the dynamics
of tidal disruption.

Speaker: Zdenek Sekanina (
Title: Tidal Breakup of the Nucleus of Comet Shoemaker-Levy 9 : A Review
The breakup of comet Shoemaker-Levy 9 is discussed both in the context of
splitting as a cometary phenomenon, comparing this object with other split
comets, and as an event with its own idiosyncrasies.  The physical appearance
of the comet is described, features diagnostic of the nature of the tidal
disruption are identified, and the implications for modelling the event
are spelled out.  Also discussed is secondary fragmentation, documenting
a continuing disintegration during 1992-94.  Much attention is paid to the
controversies that concern the models of a strengthless agglomerate versus
a~discrete cohesive mass and involve a variety of estimates for the dimensions
of the progenitor object and its major fragments.

Speaker: Wing Ip (
Title: Magnetospheric and Auroral Effects of the SL9 Impacts
As the dust has begun to settle in the preliminary analysis efforts
of the very rich set of observations of the SL9 impacts, the wide 
variety of intriguing magnetospheric phenomena observed is even more  
astounding. This is probably because the extended structures of the
dust comas of the comet fragments, and the electrodynamic coupling between 
the Jovian ionosphere and magnetosphere via electric current systems,  
could produce localized and global effects at the same times. Thus, 
magnetic field mappings from the dust comas to the Jovian atmosphere 
and from the impact sites in the southern hemisphere to their magnetic 
conjugate points in the northern hemisphere might hold the key to the 
explanation of the anomalous X-ray and ultraviolet bursts observed 
at the K and P impacts.  The time variations of the H3+ auroral emissions,
the optical and UV emissions from the Io plasma torus and the synchrotron
radiation at decimetric wavelengths could also be diagnostic of impact-related
diffusion and/or acceleration effects of magnetospheric charged particle.
In this review, an attempt will be made to provide a comprehensive overview
of the pertinent observations and their constraints imposed on theoretical 

Speaker: Emmanuel Lellouch (
Title: Chemistry Induced by the Impacts: Observations
The impacts of the comet Shoemaker-Levy 9 have produced 
considerable modifications of the composition of Jupiter's 
atmosphere at the impact sites, by (i) changing the mixing 
ratio and vertical profiles of some species originally present 
in the atmosphere (ii) generating hitherto undetected species.
In this paper, we will review spectroscopic measurements of 
minor molecular species at all wavelengths from UV to 
millimeter range. Through a comparison of these measurements, 
the goal will be to attempt establishing amounts, vertical 
profiles and if possible temporal variations, of the trace 
species, providing a basis for thermochemical, shock, and 
photochemical models of Jupiter's atmosphere after the impacts.

Speaker: Roger Yelle (
Title: Chemistry Induced by the Impacts: Theory

Speaker: Robert West (
Title: Physical & Chemical Properties and Vertical Distribution of the
        SL9 Impact Debris Particles in Jupiter's Stratosphere
Analysis of WFPC2 images from the Hubble Space Telescope suggested
that impact debris particles may owe their dark brown color to organic
material rich in S and N and that the total volume of aerosol one day
after the last impact is equal to the volume of a sphere of radius 0.5
km.  Analyses of HST and ground-based images at near-infrared
wavelengths both favor models with particle mean radius between 0.15
$\mu$m and 0.3 $\mu$m, and an aerosol spread over many scale heights,
from ~1 mb to 200 mb or greater pressure in the optically thick core
regions.  Studies of particle microphysical processes during the
fireball phase provide a relation between particle size, volatility,
and vapor concentration.  Microphysical models and observations lead
to the conclusion that the evolution of particle radius and total
optical depth during the month following the impacts was driven
chiefly by particle coagulation.  In more recent times sedimentation has
removed much of the particle mass from the stratosphere, although a
residual haze of small particles is still visible in strong near-IR
methane band images.

Speaker: Barney Conrath (
Title: Jupiter's Atmospheric Thermal Response Following 
                the Impacts of Shoemaker-Levy 9
Post-impact behavior of the atmospheric thermal structure at the 
Shoemaker-Levy 9 impact sites and surrounding regions is reviewed.  
Measurements of thermal emission in selected spectral regions in the 
mid-infrared and at millimeter wavelengths provide information on 
atmospheric temperatures between approximately 0.5 and 400 
mbar.  Based on currently available analyses, upper tropospheric 
temperatures over the sites approximately one day after impact are 
enhanced several kelvins relative to ambient, with a relaxation to 
ambient on a time scale of a few weeks.  In the stratosphere near 10 
mbar, comparable perturbations are found, but with relaxation times 
of only a few days.  In the upper stratosphere near 0.5 mbar, 
substantially larger temperature perturbations may have occurred 
with relaxation to values less than ambient, again on time scales of a 
few days.  The observed characteristic cooling times are much 
shorter than calculated ambient radiative relaxation times.  
Implications of the observed thermal structure for the dynamics 
associated with the impact sites are examined, and possible radiative 
and dynamic mechanisms responsible for the enhanced cooling are 

Speaker: Reta Beebe (
Title: Jupiter's Atmospheric Circulation from 35 to 55 Degrees
Hubble Space Telescope Wide Field Planetary Camera 2 images
obtained in the high resolution mode (240 km/ pixel) in July
1995 are used to derive zonal winds south of -30 deg.
latitude.  Eastward (E) and westward (W) maxima were
obtained at -32(W), -37(E), -39(W), -44(E), -49(W), -53(E)
and -55(W) degrees planetocentric latitude. Wind speeds
range from -20m/sec to 50 m/sec and the range of wind speeds
associated with a given jet is +/- 10 m/sec.  Sites N, H,
and L were selected as examples of small, intermediate, and
large impacts.  Mapped sequences revealed that N is drifting
with the local tropospheric winds. H is dispersing in the
winds but has an additional northward component of motion of
25+/-5 m/sec.  Material to the south of the L impact site
disperses eastward in the wind, but the east, north and west
perimeters of the site show anticyclonic circulation
consistent with a rotation rate near the outer edge of the
cloud of 45 m/sec.  This motion is independent of the local
tropospheric winds.  Maps of Q1 and C indicate that these
intermediate sites disperse in a way that is consistent with
the H site.  The K site, another large site, evolves in a
manner that is a hybrid of the H and L sites.

Speaker: Andy Ingersoll (
Title: Waves from the Impacts of SL9 with Jupiter

        A. P. Ingersoll and H. Kanamori 

Division of Geological and Planetary Sciences, California Institute 
    of Technology, Pasadena, CA 91125; e-mail:

Images of Jupiter taken by the Hubble Space Telescope (HST) reveal
circular rings surrounding five of the impact sites from comet
Shoemaker-Levy 9 (SL9).  The rings are visible 1.0 to 2.5 hours after
the impacts, and spread at a constant rate of 450 m/s (Hammel et al.,
1995).  If the material were advected by outflowing gas, it would slow
down as it moves farther from the source.  Constant spreading rate
implies that the rings are waves. However surface waves (Kanamori,
1993) and sound waves (Marley, 1994; Hunten et al., 1994) have speeds
that are too high to match the observations.  The slowest acoustic wave
is trapped in the sound channel at the tropopause temperature minimum,
and its speed is 775 m/s.  The fact that sound waves were not seen
implies that the impacts were deep.  For deep impacts the largest
amplitudes are associated with internal gravity waves trapped in a
stable layer within the clouds (Ingersoll et al., 1994).  These waves
have greater amplitude than vertically-propagating internal gravity
waves (Harrington et al., 1994).  The trapped gravity waves are
dominant even in the stratosphere, where the ring material resides.  To
match the speed, the stability of the layer must be large.  If moist
convection in the water cloud is producing the stable layer, then the
oxygen/hydrogen ratio on Jupiter is ten times that on the Sun
(Ingersoll and Kanamori, 1995). This enrichment factor for oxygen is
consistent with heavy element abundances estimated from the gravity
harmonics of Jupiter (Hubbard, 1989), but it is five times higher than
the spectroscopically-determined enrichment factors for carbon and

Speaker: Gene Shoemaker (
Title: Conference Overview
In this talk I hope to provide a summary of the key lessons learned
from the comet-Jupiter impact, as well as highlight the most important
areas that require further exploration.

sl9_icon.gifComet Shoemaker-Levy Home Page