ESO SL9 NEWS BULLETIN ===================== Issue : 12 Date : Thursday, July 21, 1994, 08:00 UT (10:00 CEST; 04:00 Chilean time) Items : 12-A: New impact parameters 12-B: The impacts of fragments P2 and N 12-C: Q less energetic than expected 12-D: Spot shapes and their changes 12-E: Spectroscopy 12-F: Press conference at ESO HQ 12-G: And here comes the next ! 12-A. NEW IMPACT PARAMETERS And then there were five... One by one, the fragments of comet SL9 disappear. Just before the final round, Paul W. Chodas and Don K. Yeomans (JPL/Caltech) have provided revised impact parameter predictions for the remaining fragments. The solutions did not change for fragments T, U, and V, as we have no new data for these. The solutions for fragments S and W were updated using Dave Jewitt's astrometric positions of July 19.3, the only new data obtained since the impacts began (item 11-F). These impact sites are closer to the limb than the earlier ones. It is clear from the observed impact phenomena that the predicted impact times have been early by an average of 5-10 minutes. This is attributed to systematic errors in the HST Guide Star Catalog, which was used to reduce most of our astrometric positions. Consistent with this is the fact that the latest data, which are reduced relative to more accurate catalogs, invariably tend to move the predicted impact times later in our solutions. As a result, it is felt that the times below are still likely to be early, but the authors refrained from applying an arbitrary correction or extrapolating the trend towards later impact times. They have, however, inflated the impact time uncertainties to account for the observed systematic errors. The predictions for all fragments are based on independent orbit solutions. 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 DE245. ------------------------------------ Fragment Impact 2-sig Angle Date/Time Unc. E-J-F July (UTC) (min) (deg) --------------h--m--s--------------- S 21 15:12:49 11.2 94.56 T 21 18:03:45 35.0 96.26 U 21 21:48:30 39.2 95.33 V 22 04:16:53 24.8 94.83 W 22 07:59:45 14.4 94.12 ------------------------------------ 12-B. THE IMPACTS OF FRAGMENTS M, N and P2 Just before issue 11 of this bulletin went to press, we received a report from the San Pedro Martir Observatory at Baja California (Mexico) about the possible observation of the impact site of the "missing" fragment M, last seen in July 1993. It seems to have been first seen at 06:13:13 UT (July 20) and appeared as a dim spot just beyond the C spot. It continued to brighten until 07:10:05 UT at which time the observers decided that "the telescope was needed for observations tomorrow (airmass >> 13)" and stopped the observations just before it hit the horizon limit switch. It was hoped that confirmation would be possible the next day, but after further calculations, it became apparent that the presumed M spot was actually the K spot. This intermezzo clearly illustrates the serious problems associated with the increasing crowding of the spots, as more and more impacts happen ! It turned out that the impact of fragment N was quite weak, as compared to the earlier ones. The Anglo-Australian-Telescope 3.9-metre Observing Team, David Crisp (JPL), Vikki Meadows (JPL), Stuart Lumsden (AAO) and Jonathan Pogson (AAO), found that the impact sites of fragments D and G were very close to the morning limb at the predicted impact time, frustrating efforts to detect the impact flash of fragment N. The seeing worsened somewhat just before the predicted impact time, with occasional periods of 3 arcsec. Nevertheless, the impact flash from fragment N was first detected on the morning limb at 10:35 UT. The intensity of this flash increased slowly until about 10:37 UT, but never exceeded 70% of the mean surface brightness of the south polar hood at 2.34 microns. At about 10:38 UT, there was an instrument problem, which lasted until about 10:42 UT. When the observations were resumed, there was no evidence of the impact site of fragment N. At 10:49 UT, IRIS was reconfigured for drift scanning with the K-grism, and the team continued to monitor the impact longitude of the site as it rotated onto the Jovian disk. There was still no evidence of the N impact site at 11:00 UT. It appears that the impact of fragment P2 was not observed. From the 2.5-metre Nordic Optical Telescope at La Palma, Pierre-Olivier Lagage writes: "We have tried to observe the P2 impact this afternoon, but at first glance nothing was detected, so that we have to analyse more thoroughly our data to have good upper limits." Kurt Birkle reports that it was not observed at Calar Alto either. At the SAAO (South Africa), Kaz Sekiguchi and Brian Carter (SAAO) obtained K-band images of Jupiter using the 0.75 m telescope and the PtSi infrared camera (designed by Ian Glass at SAAO). No obvious brightening was observed around the expected impact times of fragment P2. Doug Whittet (RPI), Dave Kilkenny (SAAO), and Jeff Shykula (RPI) obtained K-band high speed photometry around the predicted impact times of fragments P2, Q2, and Q1 using the 1.9 m telescope, but no definite brightening was recorded for P2. Nor was there any signature of impact P2 in the K band during observations at the Pic du Midi observatory (France). 12-C. Q LESS ENERGETIC THAN EXPECTED Following the spectacular impacts of fragments G, H, K and L, the expectations mounted concerning the "performance" of the brightest of the cometary fragments, Q1. The orbital calculations indicated that it would hit Jupiter at 20:04 UT, just 17 minutes after the impact of fragment Q2, which was first observed by the HST, very near the brighter Q1. Alerted by the comparatively optimistic predictions, many amateur astronomers pointed their telescopes towards Jupiter during the Q impact and in some places "Q-parties" were held. There is no doubt that the Q impact was less impressive than predicted by most astronomers. It was in fact less bright than the L impact of yesterday, one of the brightest. And in most places where observations were made, Q2 may not have been detected, at least at first glance. At the Pic du Midi observatory, a very short, tiny flash was observed in the K band at 19:44 UT, presumably due to Q2. A double flash was then observed in the K band at 20:13 and 20:19 UT from impact Q1, of which the second was the most prominent, similar in magnitude to event H from July 18. The first flash was also detected visually with telescopes of diameter 20 - 30 cm at this observatory. At the South Pole, SPIREX observed the fragment Q1 and Q2 impact sites appearing over the limb of Jupiter, although the conditions around the impact were not good, with clouds coming and going. The Q2 impact was not easily seen, due to confusion with the older impact sites. Q1 was observed at 20:20 UT, and was comparatively bright, as expected. At the SAAO, Kaz Sekiguchi and Brian Carter observed no obvious brightening around the expected impact time of Q2. The plume for fragment Q1 appeared at approximately 20:19-20:20 UT and was clearly present until about 20:30. The peak brightness was about the same as for fragment H. AT the 1.9-metre telescope, the Q1 impact plume appeared at about 20:20 UT, peaked at about 20:23:30 and returned to about the pre-impact level by 20:30 UT. As with the K-band images, low level signal detection was difficult (and any small variations seen difficult to interpret) because of the presence of bright spots. Preliminary calibrations suggest that peak brightness may have reached a K magnitude of about 1.86, as compared to 1.66 for fragment H. The French-Spanish-Swedish observing team at La Palma reported observation of the Q1 fragment impact on Jupiter at 10 microns with the Saclay CAMIRAS camera mounted on the Nordic Optical Telescope at 20:21 UT; it was less bright than the L impact seen yesterday. Soon afterwards, Mats Lindgren (Uppsala Astronomical Observatory) reported that the Q impact site had come into view. V band imaging with the Swedish solar telescope on La Palma showed it to be very similar to the nearby L site: a central dark spot and a diffuse concentric ring. The relative position between the sites now makes Jupiter look like the "Man in the Moon". The La Palma seeing was not very good, but it was just possible to see an elongation of the central dark spot, presumably due to the double impact of the Q fragments. It appears that a total of five events were observed at Calar Alto; the last of these was the strongest and most probably corresponds to the impact of fragment Q1. An image, showing the Q1 impact at 2.3 micron as imaged by the MAGIC instrument at the 3.5-metre telescope has been placed in the ESO WWW. The reasons for the less impressive effects of the Q2 and Q1 impacts may not become known before a detailed analysis has been made. An interesting comment was made by John Rogers who also reported "another excellent evening's observing at Cambridge with 30-cm refractor (date 1838, by the way). No flashes were observed visually (of course), but later there was no dark spot in the expected place for Q2, and only a rather faint, southerly dark area for Q1. The seeing was getting bad, but it seemed inferior to site H observed at the same phase two days ago. The fact that Q2 "failed" fits neatly into the pattern that all nuclei which were displaced tailward in the comet have either broken up, disappeared, or impacted without an explosion. Were these nuclei made of some tenuous tacky polymer, subject to solar wind/radiation pressure ??" Perhaps a full-scale comparative analysis of all available data from these events will provide the answer ? 12-D. SPOT SHAPES AND THEIR CHANGES Now that there are lots of "spots" in the southern hemisphere of Jupiter, the observations are entering a new phase. From many professional and recently also from amateur observatories all over the world, descriptions of the appearance of the spots and also the subtle changes are being received. A new kind of observational astronomy has emerged. We do not know how long these spots will last, and we have little idea about their developement in the turbulent Jovian atmosphere. The spots look entirely different in different wavelengths and a major observational effort is now needed to document the changes. It is fortunate that the amateur astronomers are approaching this with great enthusiasm, otherwise we could not ensure a sufficiently dense coverage in time. Now we can only wait and see - there are already some signs that some of the early spots are becoming more diffuse, but even the early A-spot, more than 80 hours after it was created, seems still to be going strong. We shall certainly hear more about this as the reports begin to come in. But it is strange to think back: who would have thought, before all this started a few days ago, that amateurs with telescopes in the 10 cm class would ever have had a chance to see any changes on Jupiter let alone participate in this exciting programme ? 12-E. SPECTROSCOPY During the past 24 hours many new reports about spectroscopic observations were received. They cover a wide variety of atoms and molecules that are studied with different instrumental techniques in several spectral regions, from ultraviolet to radio wavelengths. Although the evaluation of these data will mean a lot of hard work during the next months, some interesting indications are already apparent. Quite of few of the messages we receive over the mail exploders are dedicated to these investigations. Here is a small collection of particularly interesting examples. Matt Griffin, Andre Marten, Henry Matthews, Gary Davis, David Naylor, Greg Tompkins and Byron Han provide the following update on JCMT detection of HCN at positions of several fragments: "We have now detected HCN 4-3 at the positions of fragments C, F, G, H. We have searched for it, but not detected it from B and L. The site of fragment G impact produced the strongest line yet observed. We are monitoring the impact sites of F, G and H to examine the change in the lines over time. The strong line at G measured 21 hours after impact seems to have disappeared 35 hours after impact. Emission is also no longer observed from C. But F and/or H are still showing emission (they are 6 arcsec apart and so not unambiguously resolved by our 14 arcsec beam). We have also carried out long (0.5 hour in bad weather) integrations on the sites of fragments F and C at CO 3-2 (345 GHz), but did not detect anything." The Caltech Submillimeter Observatory has tentatively detected the 216.7 GHz (2 2 0 - 2 1 1) line of H2S at Jupiter following the fragment B impact. In the first observations of the impact site, about 1.5 hours after the predicted impact time, an extremely narrow (less than 1 MHz) absorption feature was apparently observed at the 4 sigma level. Individual scans showed that the feature, initially about 1.2 K in depth, decreased steadily to about 0.8 K over the next ten minutes. A subsequent observation of the northern hemisphere at the same frequency did not detect the line. Returning to the impact site 18 minutes after the last southern hemisphere integration, the feature had faded into the noise. Observations following the fragment C impact did not detect H2S. This may be related to the markedly different characters of the fragments B and C impacts (T. Phillips, E. Serabyn, T. Spilker and E. Weisstein). E. Lellouch, M. Festou and G. Paubert report that following the detection with the IRAM 30-metre submillimetre telescope of CO at impact site G and probably H on July 18, they have continued to monitor the CO 230 GHz emission. Re-examination of their observations of July 17 also reveals that this line was present at impact site E and marginally at impact site C. They present a detailed table of the intensities of the emissions from the different areas and their development with time, indicating a general increase, apparently beginning several hours after the impacts. The width (FWHM) of the clearly detected lines is constant at 2-3 km/s. They also mention that they have searched unsuccesfully for CO+ and H2S. John Spencer, Darren Depoy and Sang Kim, observing with the OSIRIS instrument at the CTIO 4-metre telescope report that their spectra of impact site E taken at 2:30 UT on July 18 show prominent CH4 absorption lines at 2.20, 2.30, and 2.32 microns, allowing an estimate of the cloud altitude from the CH4 abundance above the cloud layer. They have made model spectra that include CH4 absorption and solar reflection and found that the cloud deck should occur at around the 2 mbar level, assuming a normal CH4 mixing ratio in undisturbed Jupiter. However, since the explosion would bring up low altitude CH4 to the high stratosphere where CH4 mixing ratio is lower, the 2 mbar should be the lower limit of the cloud deck altitude. This is much above the cloud layers in the Jovian atmosphere: usually the NH3 cloud occurs at the 300 - 600 mbar level, and the polar haze at the 10 - 20 mbar level. NASA/IRTF reports confirmation of stratospheric ammonia from emission lines near 11.0 and 10.5 microns in a highly localized area over the impact site of fragment K at about 08:00 UT on July 20. The observations were made by the University of Texas IRSHELL spectrometer at a resolving power of 12000. A search for H2S lines near 8.10 and 8.70 microns showed no obvious emission (C. Griffith, D. Kelly, J. Lacy, G. Orton and B. Bezard for the IRTF SL9 Science Team). Observations of impacts G and H were performed with the Fourier Transform Spectrometer on CFHT on July 19 between 07:00 and 09:00 UT. The impact regions show an almost featureless continuum between 2.08 and 2.38 micron which must be responsible for the brightness of the spot seen in K filter images long after the impacts. By comparison, spectra of the same latitude on Jupiter are at zero in the same spectral range. The shape of the H2 pressure-induced absorption in these spectra implies that reflected sunlight comes from stratospheric levels of high altitudes. CO filter spectra have also been recorded on the impact G area between 06:20 and 07:00 UT showing a lower continuum, consistent with a thermal emission of Jupiter screened by a cloud layer in the impact regions (Observers: J.-P. Maillard, P. Drossart and J. Caldwell). Robert Stencel and Jessica Reynolds at the Mt. Evans-Womble Oservatory (elev. 4,313 metres, 0.6m telescope) report 1-3 micron broadband InSb observations spanning July 18.05 - 18.17 UT, including the F fragment impact area. Additional frames were recorded on July 20. Spectral scans covering the 7.7 micron methane band, 10.53 micron ethylene and 12.16 micron ethane bands were obtained July 20.20 UT. The observing efforts, particularly for Q, F and R impacts continue and comments on reduced data will follow later. Sang Kim, Christophe Dumas, Jay Elias and Richard Elston observed the southern H3+ aurora at July 18, 01:30 UT in order to measure any difference in H3+ temperature after several impacts. At that time, the B and D impact sites were visible, and it was just about 1 hour after the F impact. They used the IRS on the CTIO 1.5 m telescope with a spectral resolving power of 1600 and covered spectral range between 3.3 and 3.7 microns. They found a rotational temperature of 1000 +/- 100 K, which is very close to the normal auroral H3+ temperatures measured during the past several years. This result suggests that the SL9 impacts may not have a significant influence on the auroral activity. For the Hubble Space Telescope (HST) observing team, K. S. Noll, Space Telescope Science Institute, reports the following results: "Ultraviolet spectra of the central 0".9 of impact site G obtained with the HST Faint Object Spectrograph (FOS) on July 18 show a series of strong absorption bands over 255-293 nm that we unambiguously identify as due to S2. A second strong absorption occurs between 180 and 230 nm caused by a combination of enhanced NH3 and at least one other molecule. The absorption cross section of H2S provides a good match to the spectrum, and we regard this as a tentative identification. The absorption bands are observable in both I/F and in ratios of the impact site spectrum to the spectrum of the same location obtained on July 14. The spectral resolution of the FOS for an extended source observed through the 0".9 aperture is 0.4 nm over 160-230 nm and 0.6 nm over 230-330 nm. We note that enhancements of NH3, H2S and S2 could result from heating and vaporization of the NH4SH and NH3 clouds. If the sulphur were cometary in origin, we should also expect enhanced abundances of oxygen- and carbon-bearing compounds to be detected." (IAU Circular 6031). Moreover, "We have identified CS2 in spectra of impact site G obtained with the HST Faint Object Spectrograph. At least 13 separate bands from 1873 A to 2064 A are clearly observed. CS2 is an expected product of shock and/or solar EUV induced chemical reactions between H2S and CH4. We have also searched for bands of SO and SO2, but no band from either molecule has been observed. The two spectra were obtained at 10:43 and 11:05 UT on 19 July, more than three hours after the impact of fragment G. The 0."9 arcsec aperture was positioned at the center of the impact site." A sodium "flash" may have been observed during the H impact. Santo Catalano, a member of the Catania (Italy) group for sodium observations has reported the detection of such an event. The measurements were made with a photometer based on a Sodium Magneto-Optical Filter (SMOF) (Cacciani type), during the fragment H impact on July 18. The SMOF consists of a sodium vapour cell in a strong magnetic field and produces two narrow transmission bands (about 80 mA wide) at the neutral sodium emission lines at 5890 and 5896 A. The photometer is fed by an optical fiber placed at the Cassegrain focus of the 61 cm telescope at Serra La Nave station of Catania Astrophysical Observatory. The observations were made in a field of about 20 arcsec and show a clear enhancement of the sodium line intensity of about 15 - 20 percent. The event started at 19:57 UT and ended at 20:04 UT. It apparently occurred about 20 min after the impact, during the cooling phase of the plume. The sodium may come from the Jovian atmosphere, but observations of comets near the Sun have also shown that this element is present in comets. 12-F. PRESS CONFERENCE AT ESO HQ In view of the unexpectedly strong interest shown by the media in the Q-impacts, ESO decided to hold an informal press conference Wednesday evening at the ESO Headquarters in Garching. Beginning at 20:00 CEST, or approx. 2 hours before the impacts, the numerous participants were first informed directly by phone of the latest developments by Prof. H. Elsaesser (Director of the Max-Planck-Institute for Astronomy, Heidelberg) who was present at the German-Spanish Calar Alto observatory (Spain). During the past days, this observatory has secured several excellent series of observations of the impacts by various instruments. In particular, the rapid MAGIC infrared camera has shown the developments of several of the plumes rising over the impact sites. The weather at La Silla being so bad that observations with TIMMI could not be performed, ESO was pleased to have a running commentary by Dr. Kurt Birkle in the dome of the 3.5-metre telescope at Calar Alto during the Q-impacts. With much flair for the dramatic circumstances of this event, he described in vivid terms to the audience in Garching the atmosphere in the control room where the images kept coming in every 15 seconds. There were several smaller events, and after a while it almost seemed as if there would be no large one, contrary to all predictions. But then it happened - a small point of light appeared at the limb, it became brighter and brighter from frame to frame and the question was obvious: how high would it rise ? A certain surprise was felt when it did not get very far and it became obvious that the Q1 impact would not the gigantic event many thought it would. The astronomers at Calar Alto and in Garching were quick to point out that this particular behaviour was yet another surprise from the "comet with a thousand faces" (as somebody put it) and that this would further contribute to the understanding of the nature of comets. Clearly, at least this comet must have been put together in a very non-uniform way from very different pieces (which then behaved very differently at impact with the Jovian atmosphere). Being most certainly a left-over from the creation of the solar system, 4,600 million years ago, this must then mean that also at that time, there were many different types of bodies around. When some of these collided, they stuck to each other and what was to become comet Shoemaker-Levy 9 slowly built up in this way to become a conglomerate of different blocks. Looking further back, this new knowledge may also say something about the conditions in the interstellar cloud out of which the young solar system was formed. No more dramatic plumes rose above the Jovian limb and the astronomers declared that the Q-event was over. Tired astronomers at Calar Alto began to work on the images and to prepare one of them for the media representatives, eagerly waiting for the first Q-image in Garching. It was electronically transmitted via Heidelberg and arrived in good shape at ESO. The cameras of all the major TV channels in Germany and elsewhere recorded the emergence of the now heavily "wounded" planet on the computer screen. It did not look so faint after all: Q may not quite have reached the dimensions expected, but big it was ! Within minutes that picture was all over the world. We would like to thank Prof. Elsaesser and Dr. Birkle and all their colleagues at Calar Alto for having made this experience possible. It was another brilliant demonstration of the helpful and friendly attitude, so common among astronomers ! 12-G: AND HERE COMES THE NEXT ! Today, at 08:55 UT, the following message appeared on the screen: " We observed extremely strong CO overtone band emission in a 2.2 - 2.4 micron spectrum of the R impact at approximately July 21 05:45 UT with the CGS4 spectrometer at UKIRT, Mauna Kea, Hawaii. The CO emission faded on a timescale of 10 minutes. The overtone emission implies gas temperatures of several thousand degrees K." (Roger Knacke, Tom Geballe and Tim Brooke). ----------------- 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