Comet Hale-Bopp (C/1995 O1) : HST Observations in 1995

Our current best PRELIMINARY estimate for the effective diameter of Hale-Bopp's nucleus is:

40 kilometers (= 25 miles)

If the latter value is correct, then the nucleus of Hale-Bopp is approximately four times larger than that of comet Halley.

However, we must emphasize that measuring Hale-Bopp's size is extremely difficult, even with HST's superb spatial resolution. If you want to learn more about how we derived the above number, you can go directly to the figure from which we estimate the nuclear size . The figure caption describes what assumptions were made and why our estimate could be wrong. We have added a new figure that illustrates a little better HST's possible "photometric" detection of the nucleus.

We should also point out that there is not necessarily a direct relationship between the size of the nucleus and the strength of its activity. We learned from studying comet P/Halley and other periodic comets that, generally, only a small portion of the nuclear surface is "active". A small nucleus that is active over most of its surface area can emit more gas and dust into the coma than a much larger nucleus that is almost completely covered over by an insulating mantle of rubble. Thus, even if our preliminary estimate of the nuclear size is correct, that does not necessarily ensure that the comet will put on a spectacular visual display in early 1997.

On the other hand, a large nucleus (as indicated by the HST observations), and the huge gas and dust production rates at large heliocentric distances (as measured during the ground-based observations of CO and CN), certainly increase the odds that Hale-Bopp will not disappoint us.

Comparison of Sep 95 and Oct 95 Hale-Bopp images

Above: The above composite compares the Hubble Space Telescope (HST) images of comet Hale-Bopp taken in September (left) and October (right). In both cases the nucleus is centered in the frame, celestial North is straight up, celestial East is to the left, and each frame is 10 arcsec across. CLICK on the image to see a larger and improved version.

During September we caught Hale-Bopp only about 60 hours after an outburst, and the image shows a jet and spiral pattern emanating from the nucleus. The image from October was taken during a "quiescent" phase and barely shows any structure in the inner coma. The image in this latter case had to be strongly stretched in order to see anything (which explains why the image looks grainier than the September image). The inner coma during October is clearly elongated nearly due North. There is a very faint patch of enhanced brightness approximately 3.5 arcsec North of the nucleus, which is undoubtedly the remnant of the outburst that occurred on October 13th (i.e., 10 days prior to the HST observations).

If you like you can go directly to the October 1995 report .

Particular emphasis will be placed on results from the Hubble Space Telescope (HST) . Eleven orbits of HST observing time have been allocated to a Cycle 5 Target of Opportunity program, and the first observations were attempted during 26-27 September 1995.

DISCLAIMER

Some of the material below is based on preliminary analyses of recently acquired data. Please do not quote results from this page in any scientific or professional context without first contacting Hal Weaver . Feel free to use the information and images from this page for public outreach activities. However, these data are generally not suitable for scientific purposes without further information.

REPORT ON 26-27 SEPTEMBER 1995 HST OBSERVATIONS

HST Photo-Release Image of Hale-Bopp

Above: Hubble Space Telescope (HST) images of comet Hale-Bopp taken at ~16:00 UTC on 26 September 1995 with the WF3 CCD chip of the Wide-Field Planetary Camera 2 (WFPC2). The image to the left is from a 300 second exposure and shows how the ``pinwheel'' structure of the cometary coma clearly distinguishes Hale-Bopp from the numerous stars in the field (at the time of these images the comet was passing near the galactic plane in the constellation Sagittarius). The stars are streaked because the HST was tracking the motion of the comet. This image is 51 arcsec on each side.

The magnified view to the right is a composite created from two images, one exposed for 60 sec and the other for 300 sec, and shows more detail in the inner coma. Each pixel in this image projects to a distance of ~470 km (= 290 miles) at Hale-Bopp, and the full frame is 70 pixels (=7 arcsec) on each side. There is a ``clump'' of material about 1.4 arcsec directly above the nucleus. It appears that this material is the remnant of a piece of the nucleus that was ejected about 60 hours before this image was taken and is moving through space with a projected speed of ~30 meters per sec (=110 km per hour, or 67 miles per hour).

In both frames celestial North is at 46.7 deg CCW from the straight up direction and East is at 90 deg CCW from North. The above picture ( Postscript Version is here ) and its caption are available as Photo-Release PRC95-41 from the Office of Public Outreach of the Space Telescope Science Institute (STScI) . (Note: you should CLICK on the above image to see the larger GIF version of the photo-release.) The picture was produced with the assistance of Zolt Levay of the STScI.

Unfortunately, the HST was using an ephemeris for the comet that was offset from the true position by about 3 arcmin, but last-minute scrambling by the observing team and STScI staff succeeded in salvaging some images by uplinking pointing correction commands in real-time. The nucleus of Hale-Bopp was positioned near the center of the WF3 chip of the WFPC2 (the original plan called for putting the nucleus on the PC chip, which is the higher resolution mode of the WFPC2), and four images were successfully obtained with exposure times of 60, 300, 600, and 600 sec. The spectroscopic observations were made with a 3.66 arcsec by 1.29 arcsec aperture that was offset from the nucleus by 3 arcmin.

Incredibly, by pure coincidence we took our HST images shortly after an outburst. The comet has been monitored daily by a team of observers (led by Mark Kidger, Miquel Serra-Ricart, and Ruth Torres-Chico) using the 0.82-meter IAC-80 Telescope at the Teide Observatory in the Canary Islands. Their observations on 9/25 (about 19 hrs BEFORE the HST observations) showed no unusual activity, while observations the next night (about 5 hours AFTER the HST observations) showed a prominent jet.

The HST images seem to show a clump of dust ~1.4 arcsec from the nucleus. It appears that a substantial (TBD) chunk of material was shed by the nucleus and is disintegrating as it leaves the near-nucleus region. From the HST and Tenerife data we find that the projected velocity of the material is ~30 meters per second.

Here are some thumbnail pictures of the comet along with short text descriptions and links to larger versions of these images. ( CLICK on the thumbnail images to see the larger versions.)

Unsaturated Core Image
Above: In this stretch none of the pixels are saturated, so you can see more clearly the location of the nucleus. The peak pixel intensity is about 3.5 times brighter than the intensities in the clump of dust to the NW of the nucleus.

Question Mark Image
Above: The stretch used here makes the coma look a little like a question mark.

Clump Image
Above: The intensity stretch was chosen to emphasize the ``clump'' of dust located 1.35 arcsec directly above the nucleus, which is near the center of the frame. The peak pixel intensity at the nucleus is about 3.5 times brighter than the brightest pixels in the clump. However, the intensities in the clump are about 8 times brighter than expected for dust flowing uniformly out from the nucleus.

Log-Log Plot
Above: This plot shows spatial brightness profiles along the direction of the jet, on both sides of the nucleus. (On one side, data averaged over azimuthal angles between 87 and 121 degrees are plotted [where 0 degrees is defined as horizontal and to the right, and positive angles are measured CCW from the 0 deg direction], while on the other side data averaged over azimuthal angles between 267 and 301 degrees are plotted.) The profile for a typical comet, in which the brightness scales inversely with the projected distance to the nucleus, is also shown along with the profile for the telescope's point spread function (PSF).

Linear-Linear Plot
Above: In this plot we have divided the spatial profiles by that expected for a typical comet to show that the brightness in the clump is approximately eight times larger than expected by uniform outflow from the nucleus.

REPORT ON 23-26 OCTOBER 1995 HST OBSERVATIONS

October 1995 HST Images of Hale-Bopp

Above: Hubble Space Telescope (HST) images of comet Hale-Bopp taken at approximately 06:30 UTC on 23 October 1995 with the PC1 CCD chip of the Wide-Field Planetary Camera 2 (WFPC2). The upper left frame is a 60 sec exposure, the upper right is 300 sec long, and the lower frames are both 600 secs long. Each frame is 36.4 arcsec on a side, corresponding to 177,000 km at the comet. In all frames celestial North is at 130.7 deg CW from the straight up direction and East is at 90 deg CCW from North. (Sorry but these frames are rotated by approximately 180 deg from the September images displayed below. I'll rectify this during subsequent updates.)

These images have not been "cleaned", which means that they are littered with background stars (the comet is still near the galactic plane), cosmic ray events, and hot pixels. In the longer exposures the stars appear trailed since HST was tracking the comet. A first cut at a cleaned image appears in the report given below.

The nucleus is near the center of the frame in the highly saturated region. This particular intensity stretch was chosen in order to bring out the faint feature that's about 3.5 arcsec due north the nucleus (i.e., at 131 deg CW from straight up). This feature is almost certainly the remnant of the outburst that took place on 10/13 and which was first reported during observations on 10/14 from the Teide Observatory in the Canary Islands. The average projected speed of this clump of material is about 20 meters/sec.

The October 1995 HST observations of Hale-Bopp went beautifully. The comet was nicely centered in the PC chip (the high magnification mode of the WFPC2) during the imaging observations on October 23rd, and the comet was almost perfectly centered in the aperture of the Faint Object Spectrograph (FOS) during spectroscopic observations on October 25th.

Compared to the September observations, in which the comet was observed only about 57 hours after an outburst, the near-nucleus region was relatively dull this time around. If one uses the same intensity stretch for both the September and October data, the September image shows the now familiar spiral arm plus a clump of brightness about 1.3 arcsec from the nucleus, while the October image looks essentially like a point source. (I'll post a comparison image next week.) One has to really look hard to see the structure in the October images.

Nevertheless, upon careful inspection one can clearly see the remnants of the October 13th outburst, which was reported by the Hale-Bopp observing team at the Teide Observatory. Based on its separation from the nucleus, the clump associated with this latest outburst has a projected (on the sky) outflow speed of approximately 20 meters/sec.

We do not see any evidence for any strong outburst on October 23rd, although the inner coma is clearly elongated along the due north direction. The direction of this elongation has definitely changed significantly from its orientation in September, when it was at a position angle of 315 deg (i.e., 315 deg CCW from North).

Due to the "quiescent" nature of Hale-Bopp during these latest observations, we are hopeful that these HST images will allow us to set sensitive upper limits on the size of its nucleus. Our preliminary results are described below.

The spectroscopic data taken in October show no obvious molecular, atomic, or ionic emissions. In particular, we don't see OH, CS, CO2+, CO Cameron band emission, C, or S. We are in the process of looking carefully at these data for the purpose of setting upper limits on the above species.

Here are some thumbnail pictures of the comet along with short text descriptions and links to larger versions of these images. ( CLICK on the thumbnail images to see the larger versions.)

HST Cleaned October 1995 Hale-Bopp Image
Above: A "cleaned" image of Hale-Bopp has been produced by combining the two 600 sec exposures. Most of the star trails and other image artifacts have been removed.

October 1995 spatial brightness profile Click here (or on the image) for a magnified view.
Above: Here we compare the observed spatial brightness profile for Hale-Bopp (the "+" symbols) with a model distribution consisting of a coma (the diamond symbols) and a nucleus (the box symbols) whose diameter is 40 km. The fit seems acceptable, which is why we said above that Hale-Bopp's nucleus is roughly 40 km in diameter. However, there is more to the story.

The problem we face is the usual one in cometary research: how do we measure the light from the nucleus when it is embedded in a bright coma? Fortunately, cometary comae are usually optically thin which means that, in principle, the nucleus can be picked out provided that sufficient spatial resolution is used. Note that it is not necessary to resolve the nucleus physically (i.e., the telescope's resolution element needn't be as small as the nucleus itself); one simply needs enough resolution so that there is some contrast between light from the (unresolved) nucleus and light from the coma. As the spatial resolution is degraded, the contrast between nucleus and coma becomes more and more diluted and it becomes impossible to discern any signature of the nucleus.

The case of Hale-Bopp is particularly difficult because it is beyond the orbit of Jupiter, so that even the Hubble Space Telescope's resolution is only about 440 km (=270 miles) at the comet. Thus, in order to determine the brightness of the nucleus, we must extrapolate the coma brightness from this exterior region (i.e., outside of 440 km) down to the nucleus. If the coma light distribution is "well-behaved", then one would have some confidence that the coma brightness was well-represented in the unresolved region, allowing an accurate nuclear brightness to be determined. But if the coma shows irregularities, then any determination of the nuclear brightness is suspect. Temporal variability can cause irregularities in the coma spatial distribution, and Hale-Bopp has obviously showed strong temporal variability in the inner coma. The "knee" in the spatial profile plotted above may be due to temporal variability.

In determining the nuclear diameter, we ASSUMED that the the core of the image (i.e., the peak pixel and a few adjacent ones) was not affected by any strong outburst in activity near the time of our observations. Since the observers at the Teide Observatory reported seeing a new jet two days after our observations, we must carefully examine the assumption that our October observations were indeed characterized by "quiescent" conditions.

We also had to make an assumption about the reflectivity, or albedo, of the nucleus. We chose a geometric albedo of 4 percent, which is what was measured for the surface of comet Halley and is close to values derived for several other comets. (Albedos have been determined for only a handful of cometary nuclei, but all seem to be very dark.) The darkest known albedo for comets and asteroids is 2 percent; if Hale-Bopp's nucleus is this dark then the estimated diameter must be increased to 55 km. The largest albedo for any known planet or satellite is 90 percent; if Hale-Bopp's nucleus is this bright then the estimated diameter must be decreased to 8 km.

A major uncertainty in our analysis is our extrapolation of the coma brightness into the unresolved region. Clearly some of the observed intensity at the peak pixel is due to coma. By using the very conservative assumption that the coma is completely "flat" for the inner three points, we find that Hale-Bopp's nucleus must be smaller than 70 km (for an albedo of 4 percent).

NEW

October 1995 spatial brightness divided by a model profile Click here (or on the image) for a magnified view.
Above: Here we show another way of looking at the Hale-Bopp spatial brightness profile. We have effectively removed the rapid falloff with distance from the nucleus by dividing the data by a model coma distribution (a power law profile with an index of -1.16). The model used gives a reasonable approximation to the brightness in the inner coma, as evidenced by the fact that the ratio is approximately one (1) between 0.1 arcsec and 0.7 arcsec. The prominent dip near 0.2 arcsec is highly significant and may be due to low-level temporal variability in the comet. The deviation at large offset distances is also real and may be due to the effects of solar radiation pressure; we're not as concerned about this latter region because it does not affect our determination of the nuclear diameter.

The most important point about this figure is that the peak pixel (the point which has an x-value near zero) intensity is much brighter than would be expected for an extrapolation of the coma profile. This large deviation of the peak pixel value above the dashed line may constitute a "photometric" detection of the nucleus. Although a strong outburst in dust near the time of our observations could produce an anomalous brightening of the peak pixel, there is no evidence that any such outburst occurred. (Remember that the comet was being monitored nearly daily from the Teide Observatory in the Canary Islands.) It is also possible that the coma brightness profile steepens within approximately 400 km of the nucleus (i.e., within our resolution element; at Hale-Bopp's geocentric distance 1 PC pixel subtends 220 km at the comet), and we cannot rule that out. However, temporal variation would be the most likely cause of a steepening in the coma profile and, again, there is no evidence of any strong temporal variability near the time of our observations.

OTHER HALE-BOPP WWW SITES

o The JPL Hale-Bopp page seems to be particularly comprehensive and current.

o The ESO Hale-Bopp page was extremely active during the first month following the discovery of Hale-Bopp and continues to be the best source of information on observations conducted from the European Southern Observatory.

o The UMD Hale-Bopp Bulletin Board is an officially-designated clearinghouse for Hale-Bopp information and is brought to you by the same folks who ran a similar BB for the SL9 campaign.

o You should check out the neat IAC site , which is operated by the folks running the Tenerife Observatories in the Canary Islands. They have been observing Hale-Bopp every night using the 0.82 meter IAC-80 telescope.

o You can find finder charts and other good background information on Hale-Bopp at the Sky and Telescope Hale-Bopp page .

Comments on this WWW page can be sent to Hal Weaver via e-mail at weaver@pha.jhu.edu .

Visitors, since 3 November 1995: