ESO SL9 NEWS BULLETIN ===================== Issue : 10 Date : Tuesday, July 19, 1994, 08:00 UT (10:00 CEST; 04:00 Chilean time) Items : 10-A: La Silla observes the bright H-plume 10-B: Spectroscopic results 10-C: The powerful G-impact 10-D: Timing the impacts 10-E: Radio bursts from the G-impact 10-F: A statement 10-A. LA SILLA OBSERVES THE BRIGHT H-PLUME The observations of this unique event are becoming more and more numerous and exciting. A true barrage of email messages about new results arrive from all corners of the globe and sometimes the pace of events is simply breathtaking. There is absolutely no doubt that this observational campaign, involving astronomers all over the world, will be one of the most successful ever organised. The "poor" observers are being swamped in unique data and there is hardly any time to sit down and quietly reflect upon what they really mean. But that will of course come later when these hectic days and nights are over. Yesterday, observations commenced early at La Silla. The impact of the rather bright H-fragment was predicted to happen at 19:25 UT, that is in broad daylight in the mid-afternoon at 15:25 Chilean time. Even so, the observers at the ESO 3.6-metre telescope, Tim Livengood (NASA), Ulli Kaeufl (ESO), Benoit Mosser and Marc Sauvage (Observatoire de Paris-Meudon, France) decided to attempt observations of this event at a wavelength of 10 microns with the TIMMI instrument. It is not all that easy to point a telescope during daytime, but the 3.6-metre was ready for action, already one hour before the impact. And the reward was not long in coming. Just a few minutes later, at 19:33 UT, the H-impact was detected in the 9.1-10.4 micron band as a small point of light at the limb. It then became stronger and stronger and expanded in all directions. Only 10 - 15 minutes later, the surface brightness of the plume exceeded that of the disk of Jupiter by 50 times and the total intensity of the plume was approaching that of the entire planet. The diameter was enormous, 20,000 km, or nearly twice as big as the Earth. This impact must have involved enormous energies, so the size of the H-fragment was most certainly well above one kilometre. The 10-micron brightness persisted well over one hour, or much longer than did the A-plume. It now appears that in this waveband, the large impacts do not fade as quickly as the smaller ones do. During the following hour, about 1000 exposures were made, mostly in the 10 micron band. These show in detail the appearance of the plume and its development as it rotates into full view. The "hole" left by the G-impact (see item 10-C) is also seen near the middle of the Jovian disk. It will be interesting to see how Jupiter will look when all of the fragments have collided - how long will these new features remain visible ? Two ESO Press Photos (SL9J/94-13 and -14) were immediately prepared and placed in the ESO WWW Portal only two hours after the event. The demand was enormous and throughout the night the WWW server could hardly handle the many requests. The H-impact was observed with all other telescopes that have been allocated to the coordinated SL9 programme at La Silla. An "exclamation point" structure of the impact area was recorded by Klaus Jockers at the MPI/ESO 2.2-metre telescope and Dominique Bockelee-Morvan began a five-day programme at the SEST. The impact of fragment H was also observed from the 3.5- and 2.2-metre telescopes at Calar Alto (Spain). The initial brightening was observed at 19:31 UT and reached a maximum after about 10 minutes. It was seen at 10 microns with the CAMIRAS camera mounted on the Nordic Optical Telescope at 19:39 UT; it was much brighter than impacts A and E, previously observed with this instrument. At the South Pole, SPIREX observed the H impact site appearing over the limb at 19:38 UT. The peak brightness of the plume appeared to be smaller than the G impact, but still significantly higher than all other previous impacts. 10-B. SPECTROSCOPIC RESULTS The first results from the spectroscopic observations of the impact plumes are now beginning to appear. They show the presence of many different constituents, some of which are short-lived. There is no doubt that the impacts very effectively "stir the soup"! At La Silla, near IR spectra were recorded by Rita Schulz, Joachim A. Stuewe, Guenter. Wiedemann and Therese Encrenaz with the IRSPEC spectrometer at the 3.5-metre NTT Telescope with a 4"4 slit aligned along the parallel of the impact sites. Starting at 19:45 UT on July 18, they monitored Jupiter shortly after the impact of fragment H. A very strong emission was detected in the 3.50 to 3.56 micron range, corresponding to high levels of the methane CH4 nu3 band. The emission was monitored as it was fading rapidly away over a period of minutes. The presence of this emission indicates a high temperature (perhaps 800 - 1000 K) in the upper stratosphere; this is a clearly an effect of the impact. In addition, the H3+ multiplet at 3.5 micron was monitored between 00:00 UT and 05:00 UT on July 18 and from 20 UT to 23 UT during the folllowing evening. No evidence for H3+ emission was found near the impact regions of fragments E and F, whereas it was detected at the position of fragments D and G. On the other hand, the E-F impact region showed an emission (observed on July 18, 02:34 UT) due to the H2 S(1) line (at 2.12 micron) on top of a strong continuum. These measurements suggest that in contrast to the 2.1 micron continuum, the H3+ emission is more intense at the earlier impact sites. E. Lellouch, M. Festou and G. Paubert have observed impact sites G and H with the IRAM 30-m submillimetre telescope. They detected the CO 230 GHz line at the G impact site at 18:15 and 20:40 UT on July 18; about 25 min integration were done in each case. In both cases, the line width correspond to a turbulence of 2.5 km/sec. They also made a marginal detection of the same line at the fresh impact site H at 19:40 UT (July 18). A comparison of the intensities indicate variation of CO emission with time over a revolution after impact. The Kuiper Airborne Observatory, deployed from Melbourne, Australia, successfully detected CH4 emission at 7.7 microns associated with the G fragment using the KAO Echelle Grating Spectrograph (KEGS). C12 and C13 methane emission lines were detected at resolving power of 9000 and spatial resolution of 5 arcsec. Between approximately 07:39 and 07:52 UT (July 18) the emission increased by an order of magnitude and then steadily faded over the next 2 hours. A search was also made for Jovian water vapour at 22.6 and 23.9 microns, but a noticeable effect was not seen. This may imply that the G fragment exploded before reaching the water cloud at the 10 bar level. (Gordon Bjoraker (NASA GSFC), Terry Herter, Susan Stolovy, Bruce Pirger and George Gull (Cornell University)). Mary-Frances Jagod, Takeshi Oka, Steve Miller and Tom Geballe have identified most of the emission lines in the UKIRT CGS4 echelle spectra of Jupiter obtained the night before at the location of the C impact (approximately 15 minutes after impact). The spectra cover the range 3.53-3.56 micron and the results are as follows: 1. Six transitions of CH2 and six of OH+ are identified; 2. At least six lines remain unidentified. However, these could also be due to CH2, because the line list for this molecule is incomplete; 3. The H3+ lines present prior to impact were overwhelmed by the strong emission; even the strongest H3+ line is several times weaker than the weakest lines clearly detected in the impact spectrum, and 4. Emission by CH4 in this spectral region is ruled out. The IRTF images of the fragment C impact strongly suggest that this impact did not penetrate down to levels where water is present in Jupiter. Therefore, the identification of OH+ is a strong indication that the impactor itself contained water; this would suggest that SL9 was indeed a comet and not an asteroid. Sang Kim, Christophe Dumas, Jay Elias and Richard Elston obtained good 5 micron spectra of the D-plume with the IRS on the Cerro Tololo 1.5 m telescope around 23:15 UT (July 17). They could easily identify CH3D, CO and GeH4. The continuum intensity of the plume place was more than 10 times of those of other areas of Jupiter. Nevertheless, the overall structure of the absorption features of the plume was quite similar to those of the undisturbed areas. Several unidentified lines (probably known molecular lines) were found that must be further investigated before they can be identified. 10-C. THE POWERFUL G-IMPACT It was known from measurements of the individual fragments of SL9 that the G-fragment was the second brightest, just slightly fainter than the brightest(Q). The predicted time of impact, 09:29 CEST, was by chance just before the daily ESO Press Conference (at 11:00 CEST) when the new issue of this Bulletin is distributed. The moment of impact passed, but there was still no message to confirm that it had been observed. We had to begin to make the photocopies of Issue no. 9 and were therefore not able to bring any information about G in yesterday's issue. Consequently, the participants were once more treated to "instant astronomy". At the very moment when the conference was about to begin, there was a weak, but distinct beep from the computer terminal in the ESO auditorium, announcing the arrival of yet another email message over the network. And sure enough, there was the confirmation that the G impact had indeed been detected ! The reason for the delay was immediately evident, for the message came from a far-away place: "SPIREX detected a fragment G impact site shortly after expected impact, at 07:41 UT, significantly brighter and more persistant than any previous impacts. Details are not yet available as the South Pole communications have only been open for a few minutes. Impact G does, however, appear to be big, and extremely long lived. It has been bright for over thirty minutes. The words of Hien Nguyen, at the South Pole, were: "My God, it was extremely bright!" SPIREX will continue to monitor further impacts. (Mark Hereld, Hien Nguyen, Bernard J. Rauscher, Scott A. Severson Astronomy & Astrophysics Center, University of Chicago)." Within the minute, another message from a very different geographical location appeared on the screen: "Report from the W.M. Keck Observatory in Hawaii: The summit here in Hawaii is plagued by heavy fog. All telescopes are closed. At 07:27 UT (21:27 HST), a minute or so before the expected impact, the IRTF noticed a clearing, and we opened up. At 07:39 UT we obtained our first frame of Jupiter, in regular K-band: a truly remarkable (saturated) plume was visible well above the limb. We started a sequence of observations at 2.3 micron: A spot was visible in our first frame, 07:40 UT, which brightened to truly remarkable levels by 07:50 UT, after which it decreased in intensity. At the same time the fog was coming in, and we were closing up again; whether we were seeing a true decrease in the spot's intensity, or whether the decrease is due to increasing cloud coverage above the telescope is not yet known. It is raining right now; we don't expect to get any more frames tonight. (Imke de Pater, James Graham, Garrett Jernigan and collaborators)." And after a few more minutes came the third message from the NASA IRTF Comet Collision Team Orton, Baines, Esterle, Friedson, Goguen, Harrington, Kaminski, Lisse, Miller and Shure): "Looking through fog and 98 percent humidity, we imaged impact G at 2.29 and 4.78 +/- 0.11 (M-band) microns with a time series of 0.9 sec integrations taken every 7 sec in alternating filters. The series began at 7:35 UT, when precipitation stopped briefly, with the site already bright. The impact region increased sharply in brightness at M, eventually saturating the detector during its brightest phase. Although the 0.06"/pix images did not contain the entire planet, we estimate that the impact site outshone the planet at this wavelength. Rapidly varying fog conditions prevented a real-time analysis of the 2.29-micron images." The G-impact was indeed powerful and produced impressive effects. In Australia, the event was detected at 07:40 UT with the CASPIR infrared camera on the ANU 2.3 m telescope, and the InfraRed Imaging Spectrometer, IRIS, on the 3.9 m Anglo-Australian Telescope at Siding Spring Observatory, Coonabarabran, Australia. It occurred just as the impact sites of fragments C and A reached the evening limb of Jupiter. CASPIR 2.34-micron images and IRIS K-grism cubes showed that at 07:33 UT, the initial flash was about half as bright as the C-impact site. The brightness of this feature increased by about a factor of 4 by 07:35 UT and remained stable until about 07:40 UT. At that time, the G impact site brightened enough to saturate the detectors and to produce brilliant diffraction spikes. The AAT mirror was then stopped down to 1.9 m to prevent saturation (!). The full width of the G site was larger than the Great Red Spot (i.e., 2 to 3 times the diameter of the Earth). After 08:10 UT, the brightness of the G impact site in K-band had decreased to about 4 times the mean brightness of the fragment C site, but the G site was visible at all wavelengths where observations were made. In fact, the impact site even showed up (as a dark feature with 20,000 to 30,000 km diameter) on the TV guide camera, which is using a V filter. Heidi Hammel, on behalf of the Hubble Space Telescope Jupiter Imaging Team reported successful observations of the G plume over the limb of Jupiter. Excellent images of the G site on the disk were obtained about 1.5 hours later. Although the impact plume of nucleus G could not be observed from Japan because of clouds, the 188-cm and 91-cm Telescope Teams ( J. Watanabe, T. Yamashita, H. Hasegawa, S. Takeuchi, A. Mori, M. Abe, Y. Hirota, E. Nishihara, S. Okumura, B. Suzuki, H. Kurihara, T. Sasaki) observed the impact site when the weather again improved. A huge dark spot, larger than Red Spot was continuously observed with optical CCD camera attached to 91-cm telescope at the Okayama Astrophysical Observatory. The same feature was observed in several other observatories in Japan. Some observers said that it was easily seen, even through small telescopes. This feature was also observed as a bright spot by methane-band both in optical and near-infrared wavelength with the 188-cm telescope. CCD imaging with various filters began at the Perth Observatory from approx. 10:29 UT (evening twilight). The G impact site was just past the central meridian at the beginning of observations. The site was best seen as a bright area in the narrow (50 A) methane filter at 8930 A. Near the meridian it measured 3 1/2 arcsec in longitude (12,000 km) and had a peak surface brightness of about 1.5 times the brightness of the latitude belt that it is in. The observers in Perth, Mike A'Hearn and Dennis Wellnitz (University of Maryland) and Ralph Martin (Perth) tracked the rotation off the limb in the 8930 A filter. The site dimmed to the brightness of the belt in which it is located at 12:36:45 UT, was last detectable in an image at 12:41:00, and was totally gone at 12:42:00 UT. In the last images the feature was well above the apparent limb as seen in the deep methane band. When taken together with the observations from the South Pole at 07:41 UT, this indicated that the site is either at a very high altitude or very extended in longitiude (or both), since it was visible for more than half a Jovian rotational period. 10-D: TIMING THE IMPACTS While the most spectacular features of the present impacts on Jupiter may well be the enormous "fireballs" (plumes) which are seen with infrared-sensitive instruments, it is also possible to observe most of the impacts areas directly and some have even been sighted visually through relatively small telescopes. This, more than anything, indicates the magnitude of the disturbances in the Jovian atmosphere that is caused by these events and there are clearly great opportunities for dedicated amateur astronomers for direct participation in observations of these features. Moreover, from the accurately determined locations of the impacts, it is now possible to determine the actual moments of impacts and the difference from those predicted from the earlier orbits of the individual fragments. Quite a few descriptions are now coming over the networks about such sightings. For instance, Dan Bruton and Richard Schumde (Texas A&M Observatory) report that three dark features were observed visually using a 36 cm (f/11) Schmidt-Cassegrain Telescope. The first crossed the central meridian of Jupiter at 03:50 UT (July 18) and appeared almost like a Galilean moon shadow (roughly 6000 km in diameter). There was an irregular feature to the south-preceding edge of this dark spot. It is likely that the dark spot and irregular feature were the impact sites of F and/or E. Another feature crossed the central meridian at 04:38 UT (July 18) and appeared grey (about 9000 km in diameter); it may have been the impact site of fragment A. The third feature (transit time about 05:40 UT on July 18) was similar to the second and was probably the impact site of fragment C. Juergen Linder, a German amateur in Durmersheim, had no problem in seeing the G impact area on July 20, at 20:27 UT, near the meridian. At the Lick Observatory 120-inch telescope, Claire Max, Don Gavel, Erik Johansson (LLNL), Mike Liu (UC Berkeley) and Bill Bradford (UC, Santa Cruz) performed speckle imaging of Jupiter at 5500 A with a bare CCD camera having 0.063" pixels, from 04:25 - 05:23 on July 18. Three very dark black spots were clearly visible at the positions of impacts C, A, E (or perhaps F). Surrounding the dark spots of impacts E and A were dark partial rings or crescents. The spots were about 1 arcsec in diameter. The spot at the position of impact E (or F?) was considerably darker than that at the position of impact A at the time of observation. These dark spots were also seen at Lick by visual observers on the 36-inch refractor, as well as with CCD camera on the 40-inch Nickel telescope (Mike Brown and colleagues). They found the following central meridian crossing times and relative brightnesses: Site Age Meridian Predicted Relative (Jovian Crossing Meridian Brightness rotations) (UT) Crossing --- --------- ------- -------- ---------- E 1 03:57 03:52 18% A 3 04:46 04:39 12% C 2 05:51 05:45 16% The predicted crossing time is the one using the impact times computed by Chodas and Yeomans on July 16, and the so called System III rotation rate. These times are probably good to better than 5 minutes (the central meridian was found by astrometry from the Galilean satellites). Based on these timings, it appears that the relative times between impacts is very accurately predicted. The rotation velocity of the impact sites must be very close to the System III velocity for the timings to remain precise with 3 differently aged impact sites. Similar meridian crossing times were determined at the Palomar 5-metre telescope. Moreover, Andy Ingersoll, on behalf of the Hubble Space Telescope Imaging Team, has reported that HST has imaged impact sites A - E and accurately determined their longitudes. These longitudes were compared to the predicted longitudes by Chodas and Yeomans (last astrometric data in their solutions: 1994 July 15.0). It was found that the observed longitudes are larger (farther west) than the predicted longitudes, and that the average delay is about 10.9 minutes, with an uncertainty of about 2 minutes. The spread in the time delay is consistent with the uncertainty quoted by Chodas and Yeomans, but there appears to be a systematic delay in the observed impacts relative to the predictions. The reason for this is not yet known. 10-E: RADIO BURSTS FROM THE G-IMPACT Changes in Jupiter's radioemission were predicted when the cometary dust entered into the extended Jovian magnetosphere. Until now, however, there had been very few reports about the monitoring of this emission during the SL9 event. The only (unconfirmed) report which had reached ESO until now was from Italy where some variations had apparently been observed around the time of the first (A) impact. Now, however, a strong radio burst from Jupiter has been detected by a decameter radio receiver 26 +- 0.5 MHz at Xinxiang (China) from 06;57 to 07:07 UT on July 18. The signal was enhanced 40dBmW over the background emission (-108dBmW) of Jupiter. The signal arrived 31 minutes before the predicted impact time of fragment G. The Comet-Jupiter Impact Team of Beijing Astronomical Observatory suggests that such a strong burst is due to cyclotron emission when fragment G entered into the magnetosphere of Jupiter that has a larger extention than the atmosphere. During the same period, Koitiro Maeda (Department of Physics, Hyogo College of Medicine) and Noritaka Tokimasa and Takehiko Kuroda (Nishi-Harima Astronomical Observatory) also observed decametric radio bursts coming from Jupiter. The observations were made at 22.2 MHz with a phase-switched interferometer and a polarimeter, and also at 25.5 MHz with a radiometer. The radio bursts were polarized in the left-hand circular sense at 22.2 MHz, indicating that the radio source was presumably located in the southern polar region. The power of the burst was of the order of that of the galactic background radiation. 10-F: A STATEMENT Just as this issue was going to press (at 08:00 UT), we received the following message from Clark R. Chapman. It deserves to be made known to all involved, so we reproduce it below: "I want to put this into the historical context of Jupiter observations. It is now about 05:30 UT on July 19. I have just come in from looking at Jupiter with my back yard telescope. (Our team will again be on the Kitt Peak 2.1 m telescope three evenings from now.) The preceding end of impact site G is approximately on the central meridian. Based on my own extensive experience of observing Jupiter when I was younger, and studying historical records of Jupiter observations from the early drawings of Hooke and Cassini through the extensive 19th and 20th century reports of the British Astronomical Association, I would assert: THIS IS THE MOST VISUALLY PROMINENT DISCRETE SPOT EVER OBSERVED ON JUPITER. (By "prominence" I mean the combination of both size and contrasting albedo.) Does anyone disagree?" We do not ! And it illustrates the magnitude of the event we are now living through ! ----------------- This daily news bulletin is prepared for the media by the ESO Information Service on the occasion of the July 1994 collision between comet Shoemaker-Levy 9 and Jupiter. It is available in computer readable form over the ESO WWW Portal (URL: http://http.hq.eso.org/eso-homepage.html) and by fax to the media (on request only). News items contained therein may be copied and published freely, provided ESO is mentioned as the source. ESO Information Service European Southern Observatory Karl-Schwarzschild-Strasse 2 D-85748 Garching bei Muenchen Germany Tel.: +49-89-32006276 Fax.: +49-89-3202362