Comet Shoemaker-Levy 9

The following is a list of answers to frequently asked questions concerning the collision of comet Shoemaker-Levy 9 with Jupiter. A PostScript version and updates of this FAQ list are available at SEDS.LPL.Arizona.EDU.

General Questions


General Questions

Q1.1: Is it true that a comet will collide with Jupiter in July 1994?

Yes, the shattered comet Shoemaker-Levy 9 will collide with Jupiter over a 5.6 day period in July 1994. The first of 21 comet fragments is expected to hit Jupiter on July 16, 1994 and the last on July 22, 1994. The 21 major fragments are denoted A through W in order of impact, with letters I and O not used. All of the comet fragments will hit on the dark farside of Jupiter. The probability that all of the comet fragments will hit Jupiter is greater that 99.9%. The probability that any fragment will impact on the near side as viewed from the Earth is less than 0.01%.

The impact of the center of the comet train is predicted to occur at a Jupiter latitude of about -44 degrees at a point about 67 degrees east (toward the sunrise terminator) from the midnight meridian. These impact point estimates from Chodas and Yeomans are only 5 to 9 degrees behind the limb of Jupiter as seen from Earth. About 8 to 18 minutes after each fragment hits, the impact points will rotate past the limb. After these points cross the limb it will take another 18 minutes before they cross the morning terminator into sunlight.

Q1.2: Who are Shoemaker and Levy?

Comet Shoemaker-Levy 9 (1993e) was the ninth short-period comet discovered by Eugene and Carolyn Shoemaker and David H. Levy and was first identified on photographs taken on the night of 24 March 1993. The photographs were taken at Palomar Mountain in Southern California with a 0.46 meter Schmidt camera and were examined using a stereomicroscope to reveal the comet [2,14]. The 13.8 magnitude comet appeared 'squashed' in the original image. Subsequent photographs taken by Jim Scotti with the Spacewatch telescope on Kitt Peak in Arizona showed that the comet was actually a shattered comet. Some astronomers call the comet a "string of pearls" since the comet fragments are strung out in a line or train.

Before the end of March 1993 it was realized that the comet had made a very close approach to Jupiter in the summer of 1992. At the beginning of April 1993, after sufficient observations had been made to determine the orbit more reliably, Brian Marsden found that the comet is in orbit around Jupiter. By late May 1993 it appeared that the comet was likely to impact Jupiter in 1994. Since then, the comet has been the subject of intensive study. Searches of archival photographs have identified pre-discovery images of the comet from earlier in March 1993 but searches for even earlier images have been unsuccessful. See [11] for more information about the discovery.

Q1.3: Where can I find a GIF image of this comet?

GIF images can be obtained from SEDS.LPL.Arizona.EDU ( in the /pub/astro/SL9/images directory. Below is a list of the images at this site. The files are listed here in reverse chronological order:

There are also other collision related GIF graphics and MPEG animations at SEDS.LPL.Arizona.EDU. See Question 2.10 for other FTP and WWW sites. Most of these sites will have images of Jupiter after the collisions.

Q1.4: What will be the effect of the collision?

Jupiter will be about 770 million kilometers (480,000,000 miles) from Earth, so it will be difficult to see the effects from Earth. Also, the comet fragments will not effect Jupiter as a whole very much. It will be like sticking 21 needles into an apple: "Locally, each needle does significant damage but the whole apple isn't really modified very much." [35]. The energy deposited by the comet fragments fall well short of the energy required to set off sustained thermonuclear fusion. Jupiter would have to be more than 10 times more massive to sustain a fusion reaction.

Calculations by Paul Chodas (JPL) indicate that as seen from the Earth, the fragments will disappear behind the limb of Jupiter only 5 to 15 seconds before impact. The later fragments will be visible closer to impact. Fragment W will disappear only 5 seconds before impact, at an altitude of only about 200 km above the 1-bar pressure level: it may well start its bolide phase while still in view. Furthermore, any sufficiently dense post-impact plume will have to rise only a few hundred kilometers to be visible from Earth.

Simulations for 2-4 km fragments by Mark Boslough (Sandia National Laboratories) indicate when the fireball resulting from an impact cools it would form a debris cloud that will rise hundreds of kilometers above the Jovian cloudtops, and would enter sunlight within minutes of the impact. The arrival time of this giant cloud into sunlight would provide data on its trajectory, which in turn would help us know how big the comet fragment was. It is possible that it would be big (bright) enough to be seen by amateurs [42,43].

If the fragments are 2-4 km in diameter then the probability is very high that these effects will be visible for some of the later impacts (e.g. W and R, visible from Hawaii, S, visible from India and the far East, Q1 and Q2, visible from Africa, parts of eastern Europe and the Middle East, L, Brazil and West Africa, K, South Pacific and Australia, and maybe even V, on the final night, visible in the Western half of the U.S.). Observers in these locations are encouraged to anticipate the possibility of seeing the fireball within tens of seconds after the impact, and a few minutes later after it has cooled, condensed, and entered the sunlight. The apparent visible magnitude of any fireball will be similar to that of a Galilean satellite at best, but it would appear redder. They would be much brighter at infrared wavelengths.

The following predictions by Mordecai-Mark Mac Low (University of Chicago) are based on simulations for 1 km fragments: Each comet fragment will enter the Jupiter's atmosphere at a speed of 130,000 mph (60 km/s). At an altitude of 100 km above the visible cloud decks, aerodynamic forces will overwhelm the material strength of the fragment and tear it apart. Five seconds after entry, the comet fragment would deposit its kinetic energy of around 10^28 ergs (equivalent to around 200,000 megatons of TNT) at 100-150 km below the cloud layer [19]. Bigger fragments will have more energy and go deeper.

The hot (30,000 K) gas resulting from a 1 km stopped comet will explode, forming a fireball similar to a nuclear explosion, but much larger. The visible fireball may only rise 100 km or so above the cloudtops in this case. Above that height the density may drop so that it will become transparent. The fireball material will continue to rise, reaching a height of perhaps 1000 km before falling back down to 300 km. The fireball will spread out over the top of the stratosphere to a radius of 2000-3000 km from the point of impact. The top of the resulting shock wave will accelerate up out of the Jovian atmosphere in less than two minutes, while the fireball will be as bright as the entire sunlit surface of Jupiter for around 45 sec [18]. The fireball will be somewhat red, with a characteristic temperature of 2000 K - 4000 K (redder than the sun, which is 5800 K). Virtually all of the shocked cometary material will rise behind the shock wave, leaving the Jovian atmosphere and then splashing back down on top of the stratosphere at an altitude of 300 km above the clouds [unpublished simulations by Mac Low & Zahnle]. Not much mass is involved in this splash, so it will not be directly observable. The splash will be heavily enriched with cometary volatiles such as water or ammonia, and so may contribute to significant high hazes.

Meanwhile, the downward moving shock wave will heat the local clouds, causing them to buoyantly rise up into the stratosphere. This will allow spectroscopists to attempt to directly study cloud material, a unique opportunity to confirm theories of the composition of the Jovian clouds. Furthermore, the downward moving shock may drive seismic waves (similar to those from terrestrial earthquakes) that might be detected over much of the planet by infrared telescopes in the first hour or two after each impact. The strength of these two effects remains a topic of research. The disturbance of the atmosphere will drive internal gravity waves ("ripples in a pond") outwards. Over the days following the impact, these waves will travel over much of the planet, yielding information on the structure of the atmosphere if they can be observed (as yet an open question).

The "wings" of the comet will interact with the planet before and after the collision of the major fragments. The so-called "wings" are defined to be the distinct boundary along the lines extending in both directions from the line of the major fragments; some call these 'trails'. Sekanina, Chodas and Yeomans have shown that the trails consist of larger debris, not dust: 5-cm rock-sized material and bigger (boulder-sized and building-sized). Dust gets swept back above (north) of the trail-fragment line due to solar radiation pressure. The tails emanating from the major fragments consist of dust being swept in this manner. Only the small portion of the eastern debris trail nearest the main fragments will actually impact Jupiter, according to the model, with impacts starting only a week before the major impacts (July 8, 1994). The western debris trail, on the other hand, will impact Jupiter over a period of months following the main impacts, with the latter portion of the trail actually impacting on the front side of Jupiter as viewed from Earth [20].

The injection of dust from the wings and tail into the Jovian system may have several consequences. First, the dust will absorb many of the energetic particles that currently produce radio emissions in the Jovian magnetosphere. The expected decline and recovery of the radio emission may occur over as long as several years, and yield information on the nature and origin of the energetic particles. Second, the dust may actually form a second faint ring around the planet.

At the University of Oregon astronomers plan to monitor the linearly and circular polarized light from Jupiter. They have a sensitive detector that can detect changes of one part in 10^5 and they expect to see some kind of change in the polarization signal as the magnetic field of Jupiter is disrupted by the impacts.

Due to the great distance between Jupiter and the Earth, the comet poses no threat to Earth. However, Eugene Shoemaker says that if a similar comet crashed on Earth it would be catastrophic: "...we're talking about a million megatons of kinetic energy. We're talking about the kind of event that is associated with mass extinction of species on Earth; really and truly a global catastrophe. It might not take out the human race but it would certainly be very bad times." (CNN, Headline News)

Q1.5: Can I see the effects with my telescope?

One might be able to detect atmospheric changes on Jupiter using photography or CCD imaging. It is important, however, to observe Jupiter for several months in advance in order to know which features are due to impacts and which are naturally occurring. It appears more and more likely that most effects will be quite subtle. Without a large ( > 15" ?) telescope and good detector, little is likely to be seen.

It is possible that the impacts may create a new, temporary storm at the latitude of the impacts. Modeling by Harrington et al. suggests this is possible [30]. The fragments of comet Shoemaker-Levy 9 will strike just south of the South South Temperate Belt of Jupiter. If the nuclei penetrate deep enough, water vapor may shoot high into the atmosphere where it could turn into a bluish shroud over a portion of the South South Temperate Zone [31].

Impacts of the largest fragments may create one or two features. A spot might develop that could be a white or dark blue nodule and would likely have a maximum diameter of 2,000 km to 2,500 km which in a telescope would be 1 to 1.5 arcseconds across. This feature would be very short-lived with the impact site probably returning to normal after just a few rotations of Jupiter. A plume might also develop that would look dark against the South Temperate Zone's white clouds or could appear as a bright jet projected from Jupiter limb [31]. The table below shows the approximate sizes of features that already exist on Jupiter for comparison.

 |          FEATURE                            SIZE ESTIMATES       | 
 |       Great Red Spot                       26000 by 11000 km     |
 |     White Spots FA,BC,DE                   9000 km               |
 |  Shadows of Io, Europa, and Ganymede       4300, 4200, 7100 km   |

Below is a list of files available at in the /pub/comet directory that may be helpful in identifying features on Jupiter:      MSDOS program that displays the location of impact sites
                  and features of Jupiter
jupe.description  Description of a PC program showing features of Jupiter
jul1994.transit   Transit Times for Red Spot and White Spots for July 1994
jul1994.moons     Jovian Moon events for July 1994 (Shadows, Eclipses, etc.)

Also, there are little anticyclonic ovals at jovicentric latitudes of about -45 degrees which are typical of the South South Temperate domain and are about 3500 km in diameter. There are usually 6 or 7 around the planet and they move with the South South Temperate current, i.e. faster than BC and DE. See Sky & Telescope for a CCD image of these ovals by Don Parker [38].

Q1.6: Will there be live TV coverage of the events?

NASA's coverage of the impact of Comet P/Shoemaker-Levy 9 during the week of July 16-22 includes a series of live, televised press briefings and a 24-hour newsroom operation at the Goddard Space Flight Center (GSFC), Greenbelt, Md. The briefing panels will include Comet co-discoverers Drs. Eugene and Carolyn Shoemaker and David Levy on most days as well as scientists presenting images and information from the Hubble Space Telescope and other spacecraft. Dr. Lucy McFadden will have a round-up of observations from ground-based observatories around the world. The program and briefing schedule follows:

Sat.    16      10:00 p.m.      Live from HST: First Impact Image Release
Sun.    17      8:00 a.m.       Press Briefing at GSFC
Mon.    18      8:00 a.m.       Press Briefing at GSFC
Tue.    19      8:00 a.m.       Press Briefing at GSFC
Wed.    20      12:00 noon      Press Briefing at GSFC
Th.     21      8:00 a.m.       Press Briefing at GSFC
Fri.    22      9:30 a.m.       Press Briefing at GSFC
Sat.    23      8:00 a.m.       Press Briefing at GSFC

Note: The above times are dependent on the STS-65 mission schedule. If there is a change in the launch or landing time of the Shuttle, the program times will change. Video Uplink Schedule: NASA will provide feeds of b-roll and animation of the comet impacts with Jupiter on the following schedule: July 15, 1:00 p.m. EDT. NASA TV is carried on Spacenet 2, transponder 5, channel 9, 69 degrees West, transponder frequency is 3880 MHz, audio subcarrier is 6.8 MHz, polarization is horizontal.

CNN also plans to offer live coverage at 10:00pm EDT on July 16th with comments and explanation from scientists at Goddard and the Space Telescope Science Institute. CNN plans live coverage of at least the first 2 briefings (Sunday and Monday) and then TBA depending on what the images look like. CNN also plans to have live reports at 9am and noon EDT throughout the week with a daily comet report for release in the afternoon.

Note also that your local PBS station will normally pick up one of the live feeds listed below so that you will see it at 10:30-11:30 local time. If you have a TVRO satellite dish you can watch it 5 times.

Wednesday July 20, 1994   (into early Thursday morning 7/21)
10:30pm - 11:30pm     Telstar 401-Ku 6 (PBS Schedule B)
11:30pm - 12:30am     Telstar 401-Ku 6 (PBS Schedule B)
12:30am - 1:30 am     Telstar 401-Ku 6 (PBS Schedule B)
 1:30am - 2:30 am     Telstar 401-Ku 7 (PBS Schedule C)
 2:30am - 3:30 am     Telstar 401-Ku 5 (PBS Schedule A)
 2:30am - 3:30 am     Telstar 401-C band 8 (PBS Schedule X)

Also, there is a gathering of professional and amateur astronomers every week on the IRC (Internet Relay chat) channel #Astronomy for real time discussions. Friday and Sunday sessions are held at 20:00 UT. There may be continuous discussion/updates on the IRC channel #Astronomy during the impacts.


Q2.1: What are the impact times and impact locations?

This information was provided P.W. Chodas and D.K. Yeomans:

  Predicted Impact Parameters for Fragments of P/Shoemaker-Levy 9

  P.W. Chodas, D.K. Yeomans and Z. Sekanina (JPL/Caltech)
  P.D. Nicholson (Cornell)

  Predictions as of 1994 July 11
  Date of last astrometric data in these solutions: 1994 July 10

The predictions for all fragments except Q2 are based on independent orbit
solutions; our orbit reference identifier is given.  The orbit solution
for fragment Q2 was obtained by applying a disruption model to the orbit
for Q1, and using astrometric measurements of Q2 relative to Q1.

Except for fragment Q2, uncertainties in the impact parameters are given
immediately below the predicted values.  These uncertainties are 1-sigma
values obtained from Monte Carlo analyses; we have made an effort to make
them realistic: they are not formal uncertainty values.  NOTE: To obtain a
95% confidence level, one should use a +/- 2 sigma window around the
predicted values.  The uncertainties for Q2 have not been quantified, but
are probably comparable to those for fragment T.

The dynamical model used for these predictions includes perturbations due to
the Sun, planets, Galilean satellites and the oblateness of Jupiter.  The
planetary ephemeris used was iDE245.

Frag-    Impact       Jovicentric  Merid.  Angle          Satellite Longitudes
  ment  Date/Time     Lat.  Long.  Angle   E-J-F  Orbit     at Impact  (deg)
       July  (UTC)    (deg) (deg)  (deg)   (deg)   Ref.  Amal   Io   Eur   Gany
A = 21  16 19:57:34  -43.11  177   64.28   98.87   A20   205   345   107+   77+
               8.2      .19    5     .76     .56           4     1     1     0
B = 20  17 02:54:02  -43.15   70   63.78   99.21   B19    55+   43+  136+   91+
               6.8      .20    4     .75     .56           3     1     0     0
C = 19  17 06:59:25  -43.31  217   65.05   98.27   C16   177t   77+  153+  100+
               7.0      .17    4     .74     .55           4     1     0     0
D = 18  17 11:45:30  -43.40   29   65.47   97.95   D17   321   118+  173   11+
               7.5      .18    4     .77     .57           4     1     1     0
E = 17  17 15:05:00  -43.45  149   65.78   97.71   E33    61+  146+  187   117+
               5.6      .08    3     .46     .33           3     1     0     0
F = 16  18 00:26:39  -43.52  130   64.40   98.67   F24   343o  226   226   136+
               6.0      .12    4     .57     .41           3     1     0     0
G = 15  18 07:27:36  -43.58   23   66.58   97.10   G31   194t  286   255   151+
               4.8      .07    3     .38     .27           2     1     0     0
H = 14  18 19:25:55  -43.72   96   66.85   96.88   H30   194t   27+  305   176
               4.7      .07    3     .38     .27           2     1     0     0
K = 12  19 10:17:50  -43.79  275   67.72   96.24   K32   282   152+    8+e 207
               4.8      .07    3     .40     .28           2     1     0     0
L = 11  19 22:07:07  -43.90  343   68.09   95.95   L32   278   253    59+  232
               5.1      .07    3     .39     .28           3     1     0     0
N = 9   20 10:21:15  -44.27   67   67.87   96.03   N21   286   357o  112+  257
               6.7      .12    4     .68     .48           3     1     0     0
P2= 8b  20 15:09:51  -44.59  243   66.57   96.88   P19    71+   37+  132+  268
               6.6      .09    4     .59     .41           3     1     0     0
Q2= 7b  20 19:32:02  -44.34   39   68.83   95.33         202t   74+  150+  277
                                                           8     2     1     1
Q1= 7a  20 19:59:29  -44.04   55   69.25   95.08   Q35   216    78+  152+  278
               5.4      .07    3     .41     .29           3     1     0     0
R = 6   21 05:24:17  -44.07   36   69.35   95.00   R30   139   157   191   297
               6.2      .08    4     .47     .33           3     1     0     0
S = 5   21 15:09:53  -44.16   30   69.78   94.69   S41    74+  241   232   318
               5.9      .08    4     .42     .29           3     1     0     0
T = 4   21 18:05:50  -45.00  139   67.44   96.18   T14   161t  266   244   324
              13.2      .16    8    1.00     .70           7     2     1     0
U = 3   21 21:52:39  -44.45  274   68.98   95.20   U15   276   298   260   332
              14.5      .19    9    1.13     .79           7     2     1     1
V = 2   22 04:14:43  -44.41  145   69.32   94.96   V16   107+  352o  286   345
              10.3      .16    6    1.00     .71           5     1     1     0
W = 1   22 07:56:53  -44.16  278   70.38   94.25   W32   218    23+  302   353
               7.2      .10    4     .53     .37           4     1     1     0

Satellite Codes:  +  impact is visible from satellite
                  o  satellite is occulted by Jupiter at impact
                  e  satellite is eclipsed but not occulted at impact
                  t  satellite is in transit across Jupiter

1. Fragments J=13, M=10, and P1=8a are omitted because they have faded from
   view.  The March'94 HST images show that P2=8b and G=15 have split; we do
   not have sufficient data to obtain independent predictions for the

2. The impact date/time is the time the impact would be seen at the Earth
   (if the limb of Jupiter were not in the way); the date is the day in
   July 1994; the time is given as hours and minutes of Universal Time.
   The impact time uncertainty is a 1-sigma value in minutes.

3. The impact latitude is Jovicentric (latitude measured at the center of
   Jupiter); the Jovigraphic latitudes are about 3.84 deg more negative.

4. The impact longitude is System III, measured westwards on the planet. The
   large uncertainty in impact longitudes is due to Jupiter's fast rotation.

5. The meridian angle is the Jovicentric longitude of impact measured from
   the midnight meridian towards the morning terminator.  This relative
   longitude is known much more accurately than the absolute longitude.
   At the latitude of the impacts, the Earth limb is at meridian angle 76 deg
   and the terminator is at meridian angle 87 deg.

6. Angle E-J-F is the Earth-Jupiter-Fragment angle at impact; values greater
   than 90 deg indicate a farside impact.  All impacts will be just on the
   farside as viewed from Earth; later impacts will be closer to the limb.

7. Satellite longitudes are given for Amalthea, Io, Europa, and Ganymede.
   The longitudes are measured east from superior conjunction (the anti-Earth
   direction).  Longitude uncertainties listed as "0" are simply less than 0.5 deg.

8. According to these predictions, the only impact certain to occur during a
   satellite eclipse is K=12 with Europa eclipsed.

Since the discovery of Comet Shoemaker-Levy 9, some of the fragments of have
disappeared from view.  These were probably smaller fragments to begin with,
and they have probably disintegrated further, but some sizable pieces may
remain amongst the debris at these locations in the train.  We have computed
approximate impact times for three of these missing fragments, based on the
few available positional measurements, and using our tidal disruption model:

  Fragment    Impact      Last Seen
             July (UTC)
   J = 13    19  02:40    Dec. 1993
   M = 10    20  05:45    July 1993
   P1= 8a    20  16:30    Mar. 1994

The uncertainties on these impact times are large: probably about 60 min.,
1-sigma.  The impact locations are consistent with those of the other
fragments, i.e., at jovicentric latitude -44 and well behind the limb of
Jupiter as seen from the Earth.

[Courtesy of Paul Chodas, JPL]

     The angle of incidence of the impacts is between 41 and 42 degrees for 
all the fragments.   See "impacts.*" at SEDS.LPL.Arizona.EDU in the 
/pub/astro/SL9/info directory for updates.  The following are the 1-sigma 
(uncertainty) predictions for the fragment impact times:

               on March 1         -  30 min
               on May 1           -  24 min
               on June 1          -  16 min
               on July 1          -  10 min
               on July 15         -   6 min
               at impact - 18 hr  -   3 min

The time between impacts is thought to be known with more certainty than the
actual impact times.  This means that if somehow the impact time of the first
fragment can be measured experimentally, then impact times of the fragments
that follow can be predicted with more accuracy.

Relative Likelihood of Fireball Visibility for Fragments of Shoemaker-Levy 9
Mark Boslough and David Crawford (Sandia National Labs, Albuquerque, NM)

Fragment          Best Locations for Viewing            Likelihood of 
                   Jupiter at Impact Time            Visible Fireball (*)
   A              Africa (except W. Africa),                     :-(
                   Middle East, Eastern Europe           
   B              Eastern N. America, Mexico,                    :-(
                   Western S. America                    
   C              New Zealand, Hawaii                            :-(
   D              Australia, New Zealand,                        :-(
   E              India, Southern China,                         :-|
                   S.E. Asia, Western Australia          
   F              S. America                                     :-|
   G              New Zealand, Hawaii                            :-)
   H              Africa (except W. Africa),                     :-)
                   Middle East, Eastern Europe           
   K              Australia, New Zealand                     :-) :-)
   L              Brazil, W. Africa, Spain                   :-) :-)
   N              Australia, New Zealand                         :-|
   P2             India, Southern China,                         :-|
                   S.E. Asia, Western Australia          
   Q1             Africa (except W. Africa),             :-) :-) :-)
   Q2             Middle East, Eastern Europe                    :-)
   R              Hawaii, West coast N. America              :-) :-)
   S              India, Southern China,                 :-) :-) :-)
                   S.E. Asia, Western Australia          
   T              Africa (except W. Africa),                     :-(
                   Middle East, Eastern Europe           
   U              Brazil, W. Africa, Spain                       :-|
   V              Western U.S., Mexico                           :-|
   W              New Zealand, Hawaii,                   :-) :-) :-)
                   Eastern Australia                     
 Key:  :-) :-) :-)   Best group (Most likely to be visible)
           :-) :-)       Second best group
               :-)           Third best group
               :-|              Not very good
               :-(                 Worst (Least likely to be visible)

(*) Based on the following sources:
       1) Ballistic fireball trajectories of
          M.B. Boslough, D.A. Crawford, A.C. Robinson and T.G. Trucano
          (Sandia National Laboratories)
         "Watching for Fireballs on Jupiter", EOS, July 5, 1994.
       2) Relative brightnesses of fragments from Karen Horrocks, May, 1994.
       3) Predicted impact locations in distance beyond Jupiter's limb by
          Chodas et al. (above).

Q2.2: Can the collision be observed with radio telescopes?

The cutoff of radio emissions due to the entry of cometary dust into the Jovian magnetosphere during the weeks around impact may be clear enough to be detected by small radio telescopes. Furthermore, impacts may be directly detectable in radio frequencies. Some suggest to listen in on 15-30 MHz during the comet impact. So it appears that one could use the same antenna for both the Jupiter/Io phenomenon and the Jupiter/comet impact. There is an article in Sky & Telescope magazine which explains how to build a simple antenna for observing the Jupiter/Io interaction [4,24,25].

For those interested in radio observations during the SL9 impact, Leonard Garcia of the University of Florida has made some information available. The following files are available via anonymous ftp on the University of Florida, Department of Astronomy site in the /pub/jupiter directory:

  README.DOC      Explanation of predicted Jupiter radio storms tables
  jupradio.txt    Jovian Decametric Emission and the SL9/Jupiter Collision
  july94.txt      Tables of predicted Jupiter radio storms for July 1994

The antenna required to observe Jupiter may be as simple as a dipole antenna constructed with two pieces of wire 11 feet 8.4 inches in length, connected to a 50 ohm coax cable. This antenna should be laid out on a East-West line and raised above the ground by at least seven feet. A Directional Discontinuity Ring Radiator (DDRR) antenna is also easy to construct and can be made from 1/2 inch copper tubing 125.5 inches in length (21Mhz). The copper tube should be bent into a loop and placed 5 inches above a metallic screen. A good preamp is required for less sensitive shortwave receivers [39].

Society of Amateur and Radio Astronomers (SARA) say that amateur radio astronomers may have to wait approximately three hours after impact for the impact sites to rotate to the central meridian of Jupiter before anything unusual is detected. This wait is typical due to the Jovian decametric synchrotron emissions being emitted as a beam of radiation. Due to the large time differential from impact to radio observations any disturbance may have settled and not be detected. SARA suggest that the radio observer begin the watch approximately 30 minutes before the fragments hit to four hour after.

Q2.3: Will light from the explosions be reflected by any moons?

One may be able to witness the collisions indirectly by monitoring the brightness of the Galilean moons that may be behind Jupiter as seen from Earth. One could monitor the moons using a photometer, a CCD camera. However, current calculations suggest that the brightenings may be as little as 0.05% of the sunlit brightness of the moon [18]. If a moon can be caught in eclipse but visible from the earth during an impact, prospects will improve significantly. According to current predictions, the only impact certain to occur during a satellite eclipse is K=12 with Europa eclipsed. However, H=14 and W=1 impact only about 2 sigma after Io emerges from eclipse at longitude 20 deg, and B=20, E=17 and F=16 impact 0.5-2 sigma after Amalthea emerges from eclipse at longitude 34 deg. See also Q2.1 of this FAQ for satellite locations.

The following files contain information concerning the reflection of light by Jupiter's moons and are available at SEDS.LPL.Arizona.EDU :      MSDOS Program that Displays relative positions of
                      Jupiter's Moons during times of impact   PostScript Plot of impact times at satellite availability

Also, monitoring the eclipses of the Galilean satellites after the impacts may yield valuable scientific data with the moons serving as sensitive probes of any cometary dust in Jupiter's atmosphere. The geometry of the eclipses is such that the satellites pass through the shadow at roughly the same latitude as the predicted comet impacts. There is an article in the first issue of CCD Astronomy involving these observations. The article says that if the dust were to obscure sunlight approximately 120 kilometers above Jupiter's cloud tops, Io could be more that 3 percent (0.03 magnitudes) fainter than normal at mideclipse [40].

Q2.4: What are the orbital parameters of the comet?

Comet Shoemaker-Levy 9 is actually in a temporary orbit of Jupiter, which is most unusual: comets usually just orbit the Sun. Only two comets have ever been known to orbit a planet (Jupiter in both cases), and this was inferred in both cases by extrapolating their motion backwards to a time before they were discovered. S-L 9 is the first comet observed while orbiting a planet. Shoemaker-Levy 9's previous closest approach to Jupiter (when it broke up) was on July 7, 1992; the distance from the center of Jupiter was about 96,000 km, or about 1.3 Jupiter radii. The comet is thought to have reached apojove (farthest from Jupiter) on July 14, 1993 at a distance of about 0.33 Astronomical Units from Jupiter's center. The orbit is very elliptical, with an eccentricity of over 0.998. Computations by Paul Chodas, Zdenek Sekanina, and Don Yeomans, suggest that the comet has been orbiting Jupiter for 20 years or more, but these backward extrapolations of motion are highly uncertain. See "elements.*" and "ephemeris.*" at SEDS.LPL.Arizona.EDU in /pub/astro/SL9/info for more information.

In the abstract "The Orbit of Comet Shoemaker-Levy 9 about Jupiter" by D.K. Yeomans and P.W. Chodas (1994, BAAS, 26, 1022), the elements for the brightest fragment Q are listed. These elements are Jovicentric and for Epoch 1994Jul15 (J2000 ecliptic):

   1994 Periapses    Jul 20.7846           Arg. of periapses   43.47999
   Eccentricity      0.9987338             Long. of asc. node  290.87450
   Periapses dist.   34776.7 km            inclination         94.23333

Q2.5: Why did the comet break apart?

The comet broke apart due to tidal forces on its closest approach to Jupiter (perijove) on July 7, 1992 when it passed within the theoretical Roche limit of Jupiter. Shoemaker-Levy 9 is not the first comet observed to break apart. Comet West shattered in 1976 near the Sun [3]. Astronomers believe that in 1886 Comet Brooks 2 was ripped apart by tidal forces near Jupiter [2]. Several other comets have also been observed to have split [41].

Furthermore, images of Callisto and Ganymede show crater chains which may have resulted from the impact of a shattered comet similar to Shoemaker- Levy 9 [3,17]. The satellite with the best example of aligned craters is Callisto with 13 crater chains. There are three crater chains on Ganymede. These were first thought to be from basin ejecta; in other words secondary craters [27]. See in /pub/astro/SL9/images for images of crater chains (gipul.gif and chain.gif).

There are also a few examples of crater chains on our Moon. Jay Melosh and Ewen Whitaker have identified 2 possible crater chains on the moon which would be generated by near-Earth tidal breakup. One is called the "Davy chain" and it is very tiny but shows up as a small chain of craters aligned back toward Ptolemaeus. In near opposition images, it appears as a high albedo line; in high phase angle images, you can see the craters themselves. The second is between Almanon and Tacitus and is larger (comparable to the Ganymede and Callisto chains in size and length). There is an Apollo 11 image of a crater chain on the far side of the moon at in /pub/astro/SL9/images (moonchain.gif).

Q2.6: What are the sizes of the fragments?

Using measurements of the length of the train of fragments and a model for the tidal disruption, J.V. Scotti and H.J. Melosh have estimated that the parent nucleus of the comet (before breakup) was only about 2 km across [13]. This would imply that the individual fragments are no larger than about 500 meters across. Images of the comet taken with the Hubble Space Telescope in July 1993 indicate that the fragments are 3-4 km in diameter (3-4 km is an upper limit based on their brightness; the fragments have visual magnitudes of around 23). A more elaborate tidal disruption model by Sekanina, Chodas and Yeomans [20] predicts that the original comet nucleus was at least 10 km in diameter. This means the largest fragments could be 3-4 km across, a size consistent with estimates derived from the Hubble Space Telescope's July 1993 observations.

The new images, taken with the Hubble telescope's new Wide Field and Planetary Camera-II instrument in 1994, have given us an even clearer view of this fascinating object, which should allow a refinement of the size estimates. Some astronomers now suggest that the fragments are about 1 km or smaller. In addition, the new images show strong evidence for continuing fragmentation of some of the remaining nuclei, which will be monitored by the Hubble telescope over the next two weeks. One can get an idea of the relative sizes of the fragments by considering the relative brightnesses:

--------------------       --------------------       --------------------
          Brightness                 Brightness                 Brightness
 Nucleus    Index           Nucleus    Index           Nucleus    Index
--------------------       --------------------       --------------------
  A=21        1              J=13        0             Q1=7a        3
  B=20        1              K=12        3               R=6        2
  C=19        1              L=11        3               S=5        3
  D=18        1              M=10        0               T=4        1
  E=17        2               N=9        1               U=3        1
  F=16        2             P2=8b        2               V=2        2
  G=15        3             P1=8a        0               W=1        2
  H=14        3             Q2=7b        2
--------------------       --------------------       -------------------

The "brightness index" subjectively rates comet fragment brightnesses, 3 being brightest. Brightnesses are eyeballed from the press-released HST image where possible.

Q2.7: How long is the fragment train?

The angular length of the train was about 51 arcseconds in March 1993 [2]. The length of the train then was about one half the Earth-Moon distance. In the day just prior to impact, the fragment train will stretch across 20 arcminutes of the sky, more that half the Moon's angular diameter. This translates to a physical length of about 5 million kilometers. The train expands in length due to differential orbital motion between the first and last fragments. Below is a table with data on train length based on Sekanina, Chodas, and Yeomans's tidal disruption model:

           |    Date    Angular Length  Physical Length  |
           |               (arcsec)          (km)        |
           |  93 Mar 25       49            158,000      |
           |     Jul  1       67            265,000      |
           |  94 Jan  1      131            584,000      |
           |     Feb  1      161            669,000      |
           |     Mar  1      200            762,000      |
           |     Apr  1      255            893,000      |
           |     May  1      319          1,070,000      |
           |     Jun  1      400          1,366,000      |
           |     Jul  1      563          2,059,000      |
           |     Jul 15      944          3,593,000      |
           |    Impact A    1286          4,907,000      |

Q2.8: Will Hubble, Galileo, etc. be able to observe the collisions?

The Hubble Space Telescope, like earthlings, will not be able to see the collisions but will be able to monitor atmospheric changes on Jupiter. The impact points are favorable for viewing from spacecraft: it can now be stated with certainty that the impacts will all be visible to Galileo, and now at least some impacts will be visible to Ulysses. Although Ulysses does not have a camera, it will monitor the impacts at radio wavelengths.

Galileo will get a direct view of the impacts rather than the grazing limb view previously expected. The Ida image data playback was scheduled to end at the end of June, so there should be no tape recorder conflicts with observing the comet fragments colliding with Jupiter. The problem is how to get the most data played back when Galileo will only be transmitting at 10 bps. One solution is to have both Ulysses and Galileo record the event and and store the data on their respective tape recorders. Ulysses observations of radio emissions data will be played back first and will at least give the time of each comet fragment impact. Using this information, data can be selectively played back from Galileo's tape recorder. From Galileo's perspective, Jupiter will be 60 pixels wide and the impacts will only show up at about 1 pixel, but valuable science data can still collected in the visible and IR spectrum along with radio wave emissions from the impacts.

The impact points are also viewable by both Voyager spacecraft, especially Voyager 2. Jupiter will appear as 2.5 pixels from Voyager 2's viewpoint and 2.0 pixels for Voyager 1. However, it is doubtful that the Voyagers will image the impacts because the onboard software that controls the cameras has been deleted, and there is insufficient time to restore and test the camera software. The only Voyager instruments likely to observe the impacts are the ultraviolet spectrometer and planetary radio astronomy instrument. Voyager 1 will be 52 AU from Jupiter and will have a near-limb observation viewpoint. Voyager 2 will be in a better position to view the collision from a perspective of looking down on the impacts, and it is also closer at 41 AU.

Q2.9: To whom can I report my observations?

Observation forms by Steve Lucas are available via ftp at in the /pub/msdos/astrnomy directory. These forms also contain addresses of "Jupiter Watch Program" section leaders. contains Microsoft Write files. The Association of Lunar and Planetary Observers (ALPO) will also distribute a handbook to interested observers. The handbook "The Great Crash of 1994" is available for $10 by ALPO Jupiter Recorder, Phillip W. Budine, R.D. 3, Box 145C, Walton, NY 13856 U.S.A. The cost includes printing, postage and handling.

John Rogers, the Jupiter Section Director for the British Astronomical Association, will be collecting data from regular amateur Jupiter observers in Britain and worldwide. He can be reached via email ( or fax (UK [223] 333840). The Society of Amateur and Radio Astronomers (SARA) is collecting radio observations of the events. Observations can be sent to the chairman of the SARA Comet Watch Committee: Tom Crowley, 3912 Whittington Drive, Altanta, GA 30342, EMAIL :

Q2.10: Where can I find more information?

The SL9 educator's book put out by JPL is in the /pub/astro/SL9/EDUCATOR directory of There are two technical papers [18,19] on the atmospheric consequences of the explosions available at in the /pub/jupiter directory. There are some PostScript images and text files involving the results of fireball simulations by Sandia National Laboratories at ( in the /pub/comet/sandia directory.

SEDS (Students for the Exploration and Development of Space) has set up an anonymous account which allows you to use "lynx" - a VT100 WWW browser. To access this service, telnet to SEDS.LPL.Arizona.EDU and login as "www" (no password required). This will place you at the SEDS home page, from which you can select Shoemaker-Levy 9. A similar "gopher" interface is available at the same site. Just login as "gopher".

Anonymous ftp sites

The main links use the host names. If these don't work for you, try the IP address (number) links instead.

World Wide Web Sites


  1. "Update on the Great Comet Crash", Astronomy, December 1993, page 18.
  2. Levy, David H., "Pearls on a String", Sky & Telescope, July 1993, page 38-39.
  3. Melosh, H. H. and P. Schenk, "Split comets and the origin of crater chains on Ganymede and Callisto" Nature 365, 731-733 (1993).
  4. "Jupiter on Your Shortwave", Sky & Telescope, December 1989, page 628.
  5. "Comet on a String", Sky & Telescope, June 1993, page 8-9.
  6. "Comet Shoemaker-Levy (1993e)", Astronomy, July 1993, page 18.
  7. "A Chain of Nuclei", Astronomy, August 1993, page 18.
  8. "When Worlds Collide : Comet will Hit Jupiter", Astronomy, September 1993, page 18.
  9. Burnham, Robert "Jove's Hammer", Astronomy, October 1993, page 38-39.
  10. IAU Circulars : 5800, 5801, 5807, 5892, and 5893
  11. Observers Handbook 1994 of the R.A.S.C., Brian Marsden.
  12. Sekanina, Zdenek, "Disintegration Phenomena Expected During Collision of Comet Shoemaker-Levy 9 with Jupiter" Science 262, 382-387 (1993).
  13. Scotti, J. V. and H. J. Melosh, "Estimate of the size of comet Shoemaker-Levy 9 from a tidal breakup model" Nature 365, 733-735 (1993).
  14. Beatty, Kelly and Levy, David H., "Awaiting the Crash" Sky & Telescope, January 1994, page 40-44.
  15. Jewitt et al., Bull. Am. Astron. Soc. 25, 1042, (1993).
  16. "AstroNews", Astronomy, January 1994, page 19.
  17. "AstroNews", Astronomy, February 1994, page 16.
  18. Zahnle, Kevin and Mac Low, Mordecai-Mark, "The Collisions of Jupiter and Comet Shoemaker Levy 9", Icarus, Vol 108, page 1.
  19. Mac Low, Mordecai-Mark and Zahnle, Kevin "Explosion of Comet Shoemaker-Levy 9 on Entry into the Jovian Atmosphere", submitted to Astrophysical Journal (Letters) on 7 June 1994.
  20. Sekanina, Z., Chodas, P.W., and Yeomans, D.K, "Tidal Disruption and the Appearance of Periodic Comet Shoemaker-Levy 9", Astronomy & Astrophysics, in press.
  21. "On a collision Course with Jupiter", Mercury, Nov-Dec 1993, page 15-16.
  22. "Timing the Crash", Sky & Telescope, February 1994, page 11.
  23. "Capturing Jupiter on Video" Sky & Telescope, September 1993, page 102.
  24. "Advanced Amateur Astronomy", Gerald North, (1991), page 296-298.
  25. "Backyard Radio Astronomy", Astronomy, March 1983, page 75-77.
  26. Harrington, J., R. P. LeBeau, K. A. Backes, and T. E. Dowling, "Dynamic response of Jupiter's atmosphere to the impact of comet Shoemaker-Levy 9" Nature 368: 525-527 (1994).
  27. David Morrison, "Satellites of Jupiter", page 392, (1982).
  28. Weaver, H. A., et al, "Hubble Space Telescope Observations of Comet P/Shoemaker-Levy 9 (1993e).", Science 263, page 787-791, (1994).
  29. Duffy, T.S., W.L. Vos, C.S. Zha, H.K. Mao, and R.J. Hemley. "Sound Velocities in Dense Hydrogen and the Interior of Jupiter" Science 263, page 1590-1593, (1994).
  30. Harrington, J., R. P. LeBeau, K. A. Backes, & T. E. Dowling, "Dynamic response of Jupiter's atmosphere to the impact of comet P/Shoemaker- Levy 9", Nature 368, page 525-527, April 7, 1994.
  31. Olivares, Jose, "Jupiter's Magnificent Show", Astronomy, April 1994, page 74-79.
  32. Schmude, Richard W., "Observations of Jupiter During the 1989-90 Apparition", The Strolling Astronomer: J.A.L.P.O., Vol. 35, No. 3., September 1991.
  33. "Comet heads for collision with Jupiter",Aerospace America,April 1994, page 24-29.
  34. "Comet Shoemaker-Levy 9 and Galilean Eclipses", CCD Astronomy, Spring 1994, page 18-19.
  35. Reston, James Jr., "Collision Course", TIME, May 23, 1994, page 54-61.
  36. Benka, Stephen G., "Boom or Bust", Physics Today, June 1994, page 19-21.
  37. Beatty, Kelly and Levy, David H., "Awaiting the Crash - Part II", Sky & Telescope, July 1994, page 18-23.
  38. Alan M. MacRobert, "Observing Jupiter at Impact Time", Sky & Telescope, July 1994, page 31-35.
  39. Van Horn, Larry, "Countdown to the Crash", Monitoring Times, June 1994, page 10-13.
  40. Mallama, Anthony, "Comet Shoemaker-Levy 9 and Galilean Satellite Eclipses", CCD Astronomy, Spring 1994, pages 18-19.
  41. B. M. Middlehurst and G. P. Kuiper, "The Moon, Meteorites and Comets", Univ. Chicago Press, 1963.
  42. Boslough, Mark B., et al, "Mass and Penetration Depth of Shoemaker-Levy 9 fragments from time-resolved photometry", Geophysical Research Letters (in press), June 1994.
  43. Crawford, David A., et al, "The impact of Comet Shoemaker-Levy 9 on Jupiter", Shock Waves (in press) April 1994.
  44. Bruning, David, "The Comet Crash", Astronomy (June 1994) pages 41-45.
  45. Levy, David, "Collision course! : A comet is bearing down on Jupiter" Smithsonian, Vol 25 No 3 (June 1994) pages 62-71.
  46. Stephens, Sally, "Smash it up!", Mercury, March-Arpil 1994, pages 7-11
  47. Boslough, M.B., D.A. Crawford, A.C. Robinson, T.G. Trucano, "Watching for fireballs on Jupiter", submitted to EOS, 1994.


Thanks to Ross Smith for starting a FAQ and to all those who have contributed : Robb Linenschmidt, Mordecai-Mark Mac Low, Phil Stooke, Rik Hill, Elizabeth Roettger, Ben Zellner, Kevin Zahnle, Ron Baalke, David H. Levy, Eugene and Carolyn Shoemaker, Jim Scotti, Richard A. Schumacher, Louis A. D'Amario, John McDonald, Michael Moroney, Byron Han, Wayne Hayes, David Tholen, Patrick P. Murphy, Greg F Walz Chojnacki, Jeffrey A. Foust, Paul Martz, Kathy Rages, Paul Chodas, Zdenek Sekanina, Don Yeomans, Richard Schmude, Lenny Abbey, Chris Lewicki, the Students for the Exploration and Development of Space (SEDS), David A. Seal, Leonard Garcia, Raymond Doyle Benge, Mark Boslough, Dave Mehringer, John Spencer, Erik Max Francis, John Rogers, Al Jackson, Lucy-Ann A. McFadden, Michael F. A'Hearn, Martin Otterson, Tom Crowley, and many others who have discussed this event on newsgroups.

Dan Bruton
Physics Department
Texas A&M University
College Station, TX 77843-4242

(Formatted and hypertextified by Pat Murphy)