__________________________________________________________________________ T H E O L E R O E M E R M E S S E N G E R _______________________________________________ JJJJ C G JJJJJJ I E JJJJ __________________________________________________________________________ Newsletter of the International Jupiter Watch Satellite Discipline E-mail issue 9 May 9, 1994 Editor and Discipline Leader: John Spencer Voice: (602) 774-3358 Lowell Observatory Fax: (602) 774-6296 1400 W. Mars Hill Rd. Internet: spencer@lowell.edu Flagstaff, AZ 86001 ------------------------------------------------------------------------------ EDITORIAL COMMENTS Most of this newsletter discusses interesting developments and plans for the comet crash, but there's also some recent observational information concerning Io. ----------------------------------------------------------------------- GROUNDBASED RADIOMETRY OF IO Bob Howell (rhowell@uwyo.edu) observed an eclipse of Io by Jupiter on April 19, using the WIRO telescope in Wyoming. He writes: We observed an eclipse of Io on April 19, 1994 UT, in our special 3.4 um narrow SO2C (SO2 continuum) filter. We obtain an In-Eclipse / Out-of-Eclipse flux ratio of 0.0108 +/- 0.0019, which is somewhat less than the ratio of 0.023 obtained for the eclipse of March 4. This indicates a continued decline in activity (or continued cooling of flows) since the brightening in early 1994 reported by John Spencer. Clouds moved in shortly after the eclipse, but we do not THINK they affected the in eclipse flux. They did however prevent an absolute calibration of the data. We also obtained 5 um photometry during the May 1 - 6 period, which shows no MAJOR activity. On May 5 we observed an occultation of Io by Jupiter in the SO2C filter. Clouds confuse the early part of the event, but the latter part was clear and shows no obvious activity at Loki. Our sensitivity was such that we would not be able to detect the 1.1% hot-spot flux observed on April 19. Jay Goguen writes about his recent IRTF run, continuing the JPL program of monitoring Io's infrared lightcurve: We were on the IRTF on 4/29 and 4/30 UT with bolo and protocam. Heavy cirrus foiled any attempt at radiometry. We acquired some images of the leading hemisphere on 4/30 UT which showed no obvious hot spots. For our run, a few days after reopening, the floor refrigeration was not operating which hurts the seeing. But the seeing through the cirrus was abyssmal anyway and I wouldn't blame it all on dome seeing. So in summary, imaging of the leading side at 3.8 and 4.8 um showed no bright hot spot emission, but only a sizable outburst would have been detectable. ------------------------------------------------------------------------ OBSERVING PLANS Bob Howell will be continuing 3-5 micron speckle photometry of Io in Wyoming on June 18 - June 24. John Clarke (clarke@sunshine.sprl.umich.edu) will be obtaining FUV and visible-wavelength torus imaging of Io with HST in May 1994- see newsletter 7 for details. Nick Schneider (nick@pele.colorado.edu), Linda Sauter, and John Spencer will use CSHELL on the IRTF on May 27 and 29 to look for 1-micron emission from neutral sulfur in Io's atmosphere. Spencer will again use NSFCAM on the IRTF to monitor the Io volcanism in Jupiter eclipse on June 19 and 21. ------------------------------------------------------------------------- ISO Proposals to use the European Infrared Space Observatory (ISO), which is due to launch in late 1995, are due 19th August 1994, with letters of intent due 10th June 1994. To get the ISO INFO newsletter, or the full "Call for Observing Proposals" package, in the USA send e-mail to "iso@ipac.caltech.edu". In Europe, send to "kleech@isosa6.estec.esa.nl" ------------------------------------------------------------------------- UPCOMING CONFERENCES The Spring AGU meeting in Baltimore will have a special session on HST solar system observations, many of which concern the Jupiter system. The session will be held on Wednesday 25 May from 8:15 am - 12:00 noon. A Saturn Ring Plane Crossing Workshop will be held in Tucson on May 26-27 1994. Contact Mark Showalter (SHOWALTER@ringside.arc.nasa.gov) for more details. DPS Abstracts are due July 1st! ------------------------------------------------------------------------- PUBLICATIONS IN (OR EMERGED FROM) THE PIPELINE Submission deadline for the special issue of JGR-Planets containing papers from the Icy Galilean Satellite conference is June 1st. Here's a preprint I received that slipped through the cracks previously. Better late than never... P. Descamps Astrometric analysis of Europa-Io occultation observed in 1991. Submitted to Astronomy and Astrophysics, July 1993. ----------------------------------------------------------------------- COMET CRASH UPDATE GALILEO OBSERVATION PLANS This is from Glenn Orton in the IJW Atmospheres newsletter. Note that the Galileo PPR (Photopolarimeter/Radiometer) will obtain a direct-view lightcurve of the impact of the second fragment (B) at 0.9 microns and will send it to Earth in less than 24 hours, in time to help with the timing of ground-based observations of the later fragments. Glenn Orton expands on Clark Chapman's description of planned observations by Galileo. In the summary below, I borrow heavily from Clarke's description of the Solid-State Imaging (SSI) experiment plans and spacecraft data return strategy, but I've filled more details for the other instruments. Nearly all observations of the SL-9 impacts will be made from the Earth, or from Earth-orbiting observatories. Several spacecraft will also be observing Jupiter in mid-July. However, the Galileo spacecraft, headed for Jupiter orbit in December of 1995 will be the closest spacecraft and the only one with instrumentation capable of resolving small areas on the planet. The impact sites will be on the side of Jupiter facing the spacecraft, and Galileo's CCD camera can resolve phenomena on Jupiter as well as can be done from most ground-based observatories (Jupiter will be about 60 pixels across), at least in the 0.4- to 1.0-micron wavelength range. Other Galileo instruments will be capable of synoptic measurements at very high time resolution and at a variety of wavelengths simultaneously. Six of the 19 impacts, D, E, K, N, V, and W, will be imaged by the Solid-State Inaging (SSI) experiment, a CCD camera, in one of two basic ways. For half of the opportunities, the camera will image pictures in a time-lapse mode using a new on-chip mosaicking capability. Images will be recorded every 2-1/3 second (for fragment N) through the period that includes the bolide flash and any subsequent fireball and lasting until the impact site rotates past Jupiter's morning terminator, some 6 to 10 minutes later. Pictures will be taken every 8-2/3 seconds for two other events, V and D, which would permit successive images to be taken through a repeated cycle of 4 color filters and will reduce the 33% dead-time that characterizes the 2-1/3 second mode; this 8 2/3 second mode is more conservative of resources but would permit a brief bolide to slip through undetected, so it is best for the subsequent fireball phase. The images will be recorded on Galileo's tape recorder as arrays of 7 x 7 or 8 x 8 images per frame. A second SSI scanning mode will be most useful for recording the time history of the brief bolide flashes as the comet fragments plunge for a couple of seconds through Jupiter's upper atmosphere. Galileo's scan platform will be moved so that the image of Jupiter drifts across the CCD detector with the shutter open (through a narrow filter). The scans will sacrifice one spatial dimension, but they should measure the rise and fall of a meteor flash appoximately every 0.2 second. Diagonal scans across half of the CCD are planned for events E and W and should be sensitive to very faint phenomena, (such as meteors from fragments as small as 100 m, meteor storms, aurorae, etc.) as well as to bright phenomena. A more efficient horizontal scanning approach is planned for fragment K, but Jupiter will be superimposed on the flashes, so faint phenomena will not be detected. The Near-Infrared Mapping Spectrometer (NIMS) will observe during the impacts of C, F, G and R in a fixed-grating mode, which will record simultaneous data at some 19 wavelengths between 0.7 and 5.3 microns; these observations would provide a unique characterization of the spectrum of the impact and fireball radiation. Because the NIMS slit is only 0.5 mradians, compared with Jupiter's 0.6-mrad diameter, there is not only significant dilution of the beam, but some chance that the pointing will miss the planet. In order to avoid this possibility, the slit will be scanned across the planet in 5-1/3 sec cycles which are recorded for 2 hours on tape, a maximum of 1 hour of which might be played back. Considering the 5-1/3 sec interval as a characteristic duty cycle, then, it is possible that an impact meteor signature lasting only 2 - 3 sec might be missed. The Photopolarimeter-Radiometer (PPR) will observe during the impacts of B, H, L, P, Q, and S. Observations of P will be recorded; the remainder will be stored in a special buffer (much the same way as magnetospheric data were recorded near the Ida flyby) and transmitted during the subsequent DSN pass. The PPR field of view is 2.5 mrad circular, and, while the plume signature will be diluted, it is likely that the planet will always be in the field of view, if we try centering on the planet. As a result of this, the PPR will be set up to observe even during events without DSN coverage. Most observations will also be recorded for 40 minutes for buffer playback, ample time for undertainties in impact timing. For the P event, 80 minutes will be recorded (of which perhaps 10 min will be played back). The PPR will also observe and record during the NIMS scanning sequences as well, with Jupiter probably in its field of view most of the time. Two samples will be taken for each measurement simultaneously, allowing for cross-checking the reality of all observed events. Most observations will be done with a single filter with 0.2-sec cycle times (for which the signal-to-noise ratio for Jupiter itself is about 500). The filter usually used is centered at 945 nm, chosen to sample radiation from both the impacting meteor and the upwelling fireball. For L, the 678-nm filter will be used. For P and Q, the PPR will switch between the 678 and 945 nm filters with a 0.65-sec cycle time. The C, F, and G "ride-along" observations with NIMS will be at 945 nm; R will be at 675 nm. The Ultraviolet Spectrometer (UVS) will use its F channel, covering 162 - 323 nm with a spectral resolution of 1.36 nm, and observe simultaneously with NIMS and the PPR. The width of the UVS slit is 1.7 mrad, also larger than Jupiter's apparent disk size. The extended UVS data recorded with the P event simultaneous with the PPR will observe 3 things: (a) a "background" of Jupiter and the surrounding sky prior to the impact of the fragment, (b) the impact of P2 itself in full spectral scan mode for the F-channel (middle UV), and (c) the condensates or ejecta at the top of the atmosphere after the fireball has excavated material and when photoabsorption is taking place at the terminator. The primary objective here is to obtain a survey spectrum of the bolide flash, possibly its temperature, and ejecta composition. Even though PPR/UVS will be recording for about 1hr 20min, only about 10 minutes of data will ultimately be played back. The remaining impacts (A, K, T, U) occur during DSN playback sequences and cannot be observed by the remote sensing experiments. However the Plasma-Wave Sensor (PWS) will obtain and transmit low-rate data during all the impacts. The recorded impact data should all fit on Galileo's tape recorder, but only a few percent of it can eventually be transmitted back to Earth over Galileo's small, low-gain antenna, given the limited allocations by the Deep Space Network. In order to locate the valuable data, Galileo engineers will rely on exact timings of phenomena from Earth-based telescopic observers. They also hope to use the measurements of the first impact recorded by the PPR (B) to calibrate the timings; if the absolute timing of that impact can be measured, the relative times of the remaining impacts should be predictable to within a few minutes. The basic commands to the spacecraft will be transmitted a week or so in advance of the mid-July commencement of impacts. They will tell the spacecraft to observe for an extended time around the predicted impact times. Later information from telescopic astrometry should narrow the predicted impact times to plus-or-minus 10 minutes during the final days, and it is hoped that updated commands can be sent up to Galileo to record data for plus-or-minus 20 minutes centered on such late predictions. Galileo might even have the opportunity to change the record times for the last several impacts based on observed times for the early impacts. It is most important that the actual times of the events become known to within a couple of minutes, so that the appropriate parts of the recorded data can eventually be returned to Earth withe minimum searching for the data. The Galileo experiment will be GREATLY ENHANCED by the most rapid communication of (a) updated, more accurate pre-impact predictions of impact times from astrometry (including refinements of the stellar positions in this part of the sky derived from the ESA's Hipparcos astrometry satellite), (b) preliminary indications of when major impacts may have occurred, and (c) refined estimates of impact times from synthesis of all groundbased and spacebased data. DIRECT FIREBALL VIEWING? This also from Glenn Orton in the IJW Atmospheres newsletter: Glenn Orton notes that the proximity of the impacts to the limb (approx. 4 to 8 degrees longitude over the horizon) shows that the upwelling fireballs should be above the hroizon within about 3 minutes of the impact - still rotating into view. At this time, they should still be quite hot. If the gaseous or condensate opacities are sufficiently high to allow us to assume unit emissivity, then recent calculations of the radiance in a paper by Ahrens et al. shows that their visual magnitudes are near -2 for a 2-km diameter impactor whose density is 1 g/cm-3 and near +0.3 for a similar 400-m diameter impactor. It will be important to do more work on the opacity estimates, but the implication is now that observations of satellite flashes must be traded off against the probability of seeing the upwelling fireball directly. E-MAIL EXPLODER From the PDSSBN bulletin board, a description of plans for e-mail communications between observers during the impacts: Tests of the mail exploder for observers of the Shoemaker-Levy 9 impact will be performed in the first two weeks of May. Email sent to the exploder will be automatically relayed to everyone on the list. The list itself will be strictly limited to those conducting observations during that period and will not be publicly accessible.If you would like to be included in the SL9 exploder list, please send your name, (tentative) observing schedule, and email address at home and at the observing site to: "c1993e@astro.umd.edu". Please include "SL9 exploder" in the subject of the message. UPDATED IMPACT PREDICTIONS From Chodas and Yeomans, on the bulletin board: Here is a new edition of our Predicted Impact Parameter Table. As Jupiter is now nearing opposition, the recent astrometric data is powerful in reducing orbital uncertainties. The 1-sigma uncertainty in impact time is now 22 minutes for the major fragments. Since the orbits of some fragments are known better than others, the table now lists specific uncertainties, when available, for each fragment. The new predicted impact times are generally only half an hour later than those in our last posting, with the exception of that for fragment R, which is an hour earlier, and F, which is two hours later. The predictions for F and N are still very uncertain, but those for R are now quite reliable. The predicted impact latitudes have generally moved a few tenths of a degree northwards, mostly due to the addition of Jupiter oblateness to our dynamical model. More importantly, the predicted impact sites of the major fragments have moved by more than a degree closer to the limb as seen from the Earth, although they are all still on the farside by an average of 6 degrees. Unfortunately, the impact site for W, the closest to the limb, moved only 1/4 degree towards the limb. Monte Carlo analyses still put a very low probability on any of the impacts being on the frontside as seen from Earth. This new edition of the table includes more specific satellite data, namely satellite longitudes at impact, courtesy of Phil Nicholson. Data on Callisto has been dropped in favour of Amalthea, as the former is probably too distant to act as a useful reflector of impact flashes. Phil provides more detailed satellite data in a separate file. We use the satellite codes introduced by John Spencer in his tables. Briefly, the other changes in this edition of the table are as follows: 1) Predictions for fragment Q2 are now included. Our current orbit solution for Q2 relative to Q1 indicates that these two broke apart in the March/ April 1993 period, right around the time of discovery of Shoemaker-Levy 9. 2) The predicted time of impact is now given as hours and minutes of Universal Time; as before, the time is when the impact would be observed at the Earth. 3) The predicted System III Jovicentric longitudes of the impacts are now given, even though they are still very uncertain; the Sun-Fragment-Jupiter angle which appeared in previous tables has been deleted. 4) Perturbations due to the Galilean satellites and Jupiter oblateness have been added to our dynamical model; these improvements had only a minor effect on the impact predictions at the current level of accuracy. Paul Chodas 1994 April 26 =============================================================================== Predicted Impact Parameters for Fragments of P/Shoemaker-Levy 9 --------------------------------------------------------------- Predictions as of 1994 April 23 P.W. Chodas, D.K. Yeomans and Z. Sekanina (JPL/Caltech) P.D. Nicholson (Cornell) The predicted impact parameters for the 12 fragments E, F, G, H, K, L, N, P, Q, R, S, and W were computed from independent orbit solutions based on astrometric measurements made through April 20, 1994. The predictions for the other 8 frag-ments were computed from orbit solutions obtained by applying a tidal disruptionmodel to the orbit for fragment Q1, and using astrometric measurements relative to Q1. The method is outlined in our preprint (Sekanina, Chodas, and Yeomans). Uncertainties in the impact parameters are given immediately below the predicted values for those fragments with independent orbit solutions. These are formal 1-sigma uncertainties obtained from Monte Carlo analyses. The predictions for fragments G, H, K, L, Q, and S are the most accurate, as these have the best-known orbits; these fragments have all been assigned the same uncertainties. The impact predictions for E and W are next most accurate, followed by those for R. The predictions for F, N, and P2 are quite uncertain, as these fragments have poorly-determined orbits. The uncertainties for fragments whose orbits were computed via the tidal- disruption-model approach (A, B, C, D, Q2, T, U, and V) have not been quantified, but are probably comparable to those for fragments F and N. 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 Jovicentric Meridian Angle Satellite Longitudes Date/Time Lat. Long. Angle E-J-F at Impact (deg) July (UT) (deg) (deg) (deg) (deg) Amal Io Eur Gany ---------------h--m------------------------------------------------------------ A = 21 16 20:01 -43.31 179 65.09 98.27 207 345 107+ 77+ B = 20 17 03:11 -43.34 79 65.42 98.02 63+ 46+ 137+ 92+ C = 19 17 07:03 -43.41 218 65.59 97.88 179 78+ 154+ 100+ D = 18 17 11:58 -43.47 36 65.81 97.70 328 119+ 174 110+ E = 17 17 14:56 -43.56 144 65.40 97.96 57+ 144+ 187 116+ 27 .17 16 .76 .53 13 4 2 1 F = 16 18 02:37 -45.06 208 64.83 98.03 48+ 244 235 141+ 66 .41 40 1.80 1.30 33 9 5 2 G = 15 18 07:35 -43.71 27 66.81 96.91 197 287 256 151+ 22 .11 13 .62 .44 11 3 2 1 H = 14 18 19:23 -43.78 95 66.57 97.07 192 26+ 305 176 22 .11 13 .62 .44 11 3 2 1 K = 12 19 10:40 -43.94 287 68.25 95.83 293 155 10+e 208 22 .11 13 .62 .44 11 3 2 1 L = 11 19 21:55 -44.00 336 67.76 96.17 272 251 58+ 231 22 .11 13 .62 .44 11 3 2 1 N = 9 20 10:25 -44.97 71 65.87 97.30 288 357 o 112+ 258 60 .40 36 1.50 1.07 30 8 4 2 P2= 8b 20 15:29 -45.10 255 66.46 96.86 81+ 40+ 133+ 268 49 .29 29 1.35 .95 24 7 3 2 Q2= 7b 20 19:27 -44.38 36 68.89 95.29 200 73+ 150+ 276 Q1= 7a 20 19:54 -44.10 52 69.31 95.04 213 77+ 152+ 277 22 .11 13 .62 .44 11 3 2 1 R = 6 21 05:41 -44.20 46 69.61 94.80 148 160 193 298 33 .16 20 .88 .63 16 5 2 1 S = 5 21 15:24 -44.18 38 70.20 94.38 80+ 243 233 318 22 .11 13 .62 .44 11 3 2 1 T = 4 21 18:30 -44.23 150 70.30 94.30 174 269 246 325 U = 3 21 21:43 -44.26 267 70.44 94.19 271 297 259 332 V = 2 22 04:48 -44.27 164 70.75 93.96 124+ 357 o 289 346 W = 1 22 08:19 -44.21 290 71.05 93.77 229 26+ 303 354 27 .17 16 .76 .53 13 4 2 1 Satellite Codes: + impact is visible from satellite o satellite is occulted by Jupiter at impact e satellite is eclipsed but not occulted at impact ------------------------------------------------------------------------------- Notes: 1. Fragments J=13 and M=10 are omitted from the Table because they have faded from view. Fragments P=8 and Q=7 each consist of two major components: P2=8b is the brighter and more easterly of the two P components in the January'94 HST image, while Q1=7a is the brighter and more southerly of the two Q fragments. The March'94 HST image shows that P1=8a has almost completely faded away (so it too is omitted from the Table), and that P2=8b has split. We do not as yet have sufficient data to obtain independent predictions for the two components of P2=8b. 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 1-sigma uncertainties, when shown, are 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. Although System III applies to the interior of the planet, not the atmosphere, it is more deterministic than Systems I or II. The large uncertainty in impact longitudes is due to the uncertainty in the impact times and 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. 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; the later impacts will be closest to the limb. According to our Monte Carlo analyses, the probability that any fragment will impact on the near side as viewed from the Earth is < 0.01%. 7. Satellite data are given for Amalthea, Io, Europa, and Ganymede. Callisto is omitted, as it is too distant to act as a useful reflector of the impact flashes, and it has no occultations or eclipses during the impacts. Metis, Adrastea and Thebe are also omitted, due to the expected faintness of any flash reflections from them. The satellite longitudes are measured east from superior conjunction (the anti-Earth direction). 8. According to these 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. Additional note from Phil Nicholson: Estimated K magnitudes (wavelength 2.2 microns) for each satellite in sunlight and in eclipse but lit by an impact with a peak luminosity of 1.2e25 erg/s are: ------------------------------------------ Flash K Magnitude % Brightening Satellite Sunlight Eclipse in sunlight ------------------------------------------ Metis 15.1 17.2 15 Adrastea 16.5 18.6 15 Amalthea 12.2 15.0 7 Thebe 14.0 17.3 5 Io 3.3 7.7 1.3 Europa 2.6 9.0 0.5 Ganymede 3.3 9.3 0.2 Callisto 3.6 -- 0.1 ------------------------------------------