Icarus 179 (2005) 174–194 www.elsevier.com/locate/icarus
Physical characteristics of Comet Nucleus C/2001 OG108 (LONEOS) Paul A. Abell a,∗,1,2 , Yanga R. Fernández b,3 , Petr Pravec c , Linda M. French d,4 , Tony L. Farnham e , Michael J. Gaffey f,1 , Paul S. Hardersen f,1 , Peter Kušnirák c , Lenka Šarounová c , Scott S. Sheppard g , Gautham Narayan d a Planetary Astronomy Group, Astromaterials Research and Exploration Science, NASA Johnson Space Center, Mail Code KR,
Houston, TX 77058-3696, USA b Institute for Astronomy, University of Hawai’i, Honolulu, HI 96822, USA c Astronomical Institute, Academy of Sciences of the Czech Republic, CZ-25165, Ondˇrejov, Czech Republic d Department of Physics, Illinois Wesleyan University, Bloomington, IL 61702, USA e Department of Astronomy, University of Maryland, College Park, MD 20742, USA f Department of Space Studies, University of North Dakota, Grand Forks, ND 58202, USA g Department of Terrestrial Magnetism, Carnegie Institution of Washington, Washington, DC 20015, USA
Received 26 September 2004; revised 28 May 2005 Available online 24 August 2005
Abstract A detailed description of the Halley-type Comet C/2001 OG108 (LONEOS) has been derived from visible, near-infrared, and mid-infrared observations obtained in October and November 2001. These data represent the first high-quality ground-based observations of a bare Halley-type comet nucleus and provide the best characterization of a Halley-type comet other than 1P/Halley itself. Analysis of time series photometry suggests that the nucleus has a rotation period of 57.2 ± 0.5 h with a minimum nuclear axial ratio of 1.3, a phase-darkening slope parameter G of −0.01 ± 0.10, and an estimated H = 13.05 ± 0.10. The rotation period of C/2001 OG108 is one of the longest observed among comet nuclei. The V -R color index for this object is measured to be 0.46 ± 0.02, which is virtually identical to that of other cometary nuclei and other possible extinct comet candidates. Measurements of the comet’s thermal emission constrain the projected elliptical nuclear radii to be 9.6 ± 1.0 km and 7.4 ± 1.0 km, which makes C/2001 OG108 one of the larger cometary nuclei known. The derived geometric albedo in V -band of 0.040 ± 0.010 is typical for comet nuclei. Visible-wavelength spectrophotometry and near-infrared spectroscopy were combined to derive the nucleus’s reflectance spectrum over a 0.4 to 2.5 µm wavelength range. These measurements represent one of the few nuclear spectra ever observed and the only known spectrum of a Halley-type comet. The spectrum of this comet nucleus is very nearly linear and shows no discernable absorption features at a 5% detection limit. The lack of any features, especially in the 0.8 to 1.0 µm range such as are seen in the spectra of carbonaceous chondrite meteorites and many low-albedo asteroids, is consistent with the presence of anhydrous rather than hydrous silicates on the surface of this comet. None of the currently recognized meteorites in the terrestrial collections have reflectance spectra that match C/2001 OG108 . The near-infrared spectrum, the geometric albedo, and the visible spectrophotometry all indicate that C/2001 OG108 has spectral properties analogous to the D-type, and possibly P-type asteroids. Comparison of the measured albedo and diameter of C/2001 OG108 with those of Damocloid asteroids reveals similarities between these asteroids and this comet nucleus, a finding which supports previous dynamical arguments that Damocloid asteroids could be composed of cometary-like materials. These observations are also consistent with findings that two Jupiter-family comets may have spectral signatures indicative of D-type asteroids. C/2001 OG108 probably represents the transition from a typical active comet to an extinct cometary nucleus, and, as a Halley-type comet, * Corresponding author. Fax: +1 281 483 5276.
E-mail address:
[email protected] (P.A. Abell). 1 Visiting astronomer at the Infrared Telescope Facility, which is operated by the University of Hawai’i under Cooperative Agreement no. NCC 5-538 with
the National Aeronautics and Space Administration, Office of Space Science, Planetary Astronomy Program. 2 National Research Council Associate. 3 Visiting astronomer at the W.M. Keck Observatory, which is jointly operated by the California Institute of Technology and the University of California. 4 Visiting astronomer at Lowell Observatory. 0019-1035/$ – see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.icarus.2005.06.012
Characteristics of C/2001 OG108 (LONEOS)
175
suggests that some comets originating in the Oort cloud can become extinct without disintegrating. As a near-Earth object, C/2001 OG108 supports the suggestion that some fraction of the near-Earth asteroid population consists of extinct cometary nuclei. 2005 Elsevier Inc. All rights reserved. Keywords: Comets, composition; Infrared observations; Visible observations; Comets, origin; Near-Earth asteroids, origin
1. Introduction Cometary nuclei represent objects which have remained relatively pristine since the early formation of the Solar System. Due to their large heliocentric formation distances (>4 AU), they probably have not been as thermally processed as those objects residing in the inner asteroid belt, and are more likely to contain a significant amount of volatiles (Wyckoff, 1982). In addition, they may have sampled entirely different reservoirs of material that condensed from the late solar nebula than their asteroid counterparts, which formed in the innermost regions of the Solar System. Therefore, observations of cometary nuclei could give some insights into the probable material species that existed in these outer regions of the early Solar System, giving investigators a better understanding of the possible physical properties of these objects, and a more complete inventory of the distribution of materials within the early Solar System (Kidger, 2002). Such observations should not only provide clues about the possible compositional characteristics of comets, but also provide information on their evolutionary history, and potential relationship to other bodies, such as interplanetary dust particles, meteorites, and asteroids. For the past few decades, scientists have been exploring possible connections between specific populations of Solar System bodies and cometary nuclei. One such group of objects that has been the focus of this kind of investigation is the near-Earth asteroid population. These asteroids have been recognized as some of the more scientifically interesting objects in the Solar System because, unlike the orbits of asteroids in the mainbelt population, near-Earth asteroid orbits are not stable over the 4.6 billion-year age of the Solar System (Öpik, 1951). Their dynamical lifetimes are only on the order of ∼106 to 108 years due to interactions with other objects in the inner Solar System that cause them to either impact one of the inner planets or the Sun, or be ejected from the Solar System altogether (Morbidelli and Gladman, 1998). Hence the presence of these bodies requires a mechanism(s) and source region(s) to replenish and maintain the near-Earth asteroid population over time. The mainbelt asteroids have been recognized as one of the primary sources of material for the near-Earth asteroid population (McFadden et al., 1985; Morbidelli et al., 2002), but several investigators have suggested that a nonnegligible portion of the near-Earth asteroid population could also be replenished by cometary nuclei that have evolved dynamically into the inner Solar System from such reservoirs as the Edgeworth–Kuiper belt and the Oort
cloud (Öpik, 1961, 1963; Wetherill, 1971; Kresák, 1979; Shoemaker et al., 1979; Degewij and Tedesco, 1982; Weissman et al. 1989, 2002; Harris and Bailey, 1998). Evidence used to support the hypothesis of a cometary component to the near-Earth asteroid population was based on: observations of asteroid orbits and associated meteor showers (e.g., 3200 Phaethon and the Geminid meteor shower) (Whipple, 1983; Fox et al., 1984; Olsson-Steel, 1988; Williams and Wu, 1993); low activity of shortperiod comet nuclei, which implied nonvolatile surface crusts (e.g., 28P/Neujmin 1, 49P/Arend-Rigaux) (A’Hearn, 1988); and detection of possible transient cometary activity in a near-Earth asteroid (e.g., 4015 Wilson–Harrington) (Cunningham, 1950; Bowell et al., 1992; McFadden, 1993; Fernández et al., 1997). Previous dynamical studies have concluded that as much as 40–50% of the near-Earth asteroid population could be due to extinct comets (Wetherill, 1988, 1991; Binzel et al., 1992), but more recent investigations based on physical and dynamical evidence have suggested that approximately 5–10% of the near-Earth asteroid population may be extinct comets with the remaining fraction made up of fragments from parent bodies within the mainbelt asteroid population (Fernández et al., 2001; Bottke et al., 2002). The uncertainty of the cometary contribution to the nearEarth asteroid population is partly due to the lack of an observational discriminator that will distinguish between an extinct comet and a “true” asteroid. In addition, the observational techniques successfully used to study asteroid surfaces are usually not applicable to comets because their comae dominate any signal from the nucleus during their perihelion passages (Weissman et al., 2002). In order to examine known comet surfaces directly, it is either necessary to study them during their quiescent phase, when they are at large heliocentric distances and thus extremely faint (i.e., low signal-tonoise), or to develop spacecraft missions to rendezvous with them. Both of these observational strategies have produced results, but only a few cometary nuclei have been adequately studied in enough detail to constrain their physical characteristics and compositions through near-infrared spectroscopy (Soderblom et al., 2002; Licandro et al., 2002, 2003). However, if a significant fraction (∼5–10%) of the nearEarth asteroid population is actually composed of extinct cometary nuclei, there should be some objects within this subset of the population that demonstrate low-levels of coma. These low-activity comets may represent objects undergoing the transition from active comets to extinct cometary nuclei. Such objects may have been nearly de-
176
P.A. Abell et al. / Icarus 179 (2005) 174–194
pleted of their entire volatile content, or at least their nearsurface volatiles, leaving behind a nonvolatile and poorly conducting surface crust that could prevent solar insolation from sublimating subsurface ice (Whipple, 1950; Weissman, 1980; A’Hearn, 1988; Levison and Duncan, 1994, 1997; Weissman and Levison, 1997; Weissman et al., 2002). Coma from these objects would be frequently undetectable or nonexistent, and these comet nuclei would resemble asteroids. Therefore, low-activity comets may help constrain the estimated contribution of cometary objects to the near-Earth asteroid population and lead to a better understanding of the mechanisms related to the last stages of cometary evolution. The detections of coma from near-Earth object 2001 OG108 , and its subsequent re-classification from an asteroid to a comet, have therefore renewed interest in determining the physical properties of comet–asteroid transition objects. Near-Earth object 2001 OG108 was discovered by the Lowell Observatory Near Earth Asteroid Search (LONEOS) program (described by Stokes et al., 2002), on July 28, 2001. Marsden (2001) soon after published an orbit for the object with a 50-year period and an inclination almost perpendicular to the ecliptic. It was quickly apparent that this object had an orbit similar to the Halley-type comets and thus was a member of the so-called “Damocloids,” i.e., asteroids like (5335) Damocles on high-inclination and large semimajor axis orbits (Asher et al., 1994; Bailey and Emel’yanenko, 1996; Bowell, 2001). Almost 20 Damocloids are now known. Subsequent astrometry of 2001 OG108 over the next few months refined the orbit (Marsden, 2002) to a period of 48.5 years, semi-major axis of 13.31 AU, eccentricity of 0.925, and orbital inclination of 80.2 degrees. Due to its orbital similarity to the Damocloid asteroids and Halley-type comets, it was thought that 2001 OG108 might possibly be an inactive or dormant comet nucleus. Therefore, several groups began monitoring the object as it proceeded toward perihelion in the hopes that it would show signs of cometary activity. However, examination of the initial discovery observations and those obtained several months later showed no evidence of coma (Marsden, 2001; French, 2002). Hence the asteroid classification of 2001 OG108 remained. It was not until observations were obtained during January and February 2002, which showed that the object had developed a slight amount of coma as it approached perihelion (Nakamura et al., 2002), that 2001 OG108 was re-classified and designated as comet C/2001 OG108 (LONEOS) (Marsden, 2002). All the data presented in this paper are based on multiple studies that took place during October and November 2001, while the object was still relatively bright, but before any coma was detected (French, 2002). C/2001 OG108 ’s first report of activity was roughly at a heliocentric distance of 1.4 AU in January 2002. This activity continued through May 2002, and stopped when the comet was at heliocentric distances of approximately 1.4 to 1.5 AU (Filonenko and
(a)
(b) Fig. 1. (a) A 120-second R band image of C/2001 OG108 (LONEOS) taken with the Lowell Observatory 1.1-m Hall telescope on October 23, 2001 at 03:30 UT. The comet is located next to the white arrow in this image. North is up and east is to the left. Linear width of the image at the position of the comet is approximately 292,000 km (277.2 arcsec). Note that no coma is evident from the object in this image. (b) A 1200-second R band image of C/2001 OG108 (LONEOS) taken with the University of Hawai’i 88-inch telescope on June 1, 2002 at 06:45 UT. North is up and east is to the left. Linear width of the image at the position of the comet is approximately 75,000 km (76.1 arcsec).
Churyumov, 2003). The appearance of the comet before and after activity is shown in Fig. 1a and 1b. The core of the comet is point-like in each case, indicating the absence of a “canonical” cometary coma. There may be some lingering cometary tail in the 2002 image (Fig. 1b), but the comet’s photocenter mimics the point-spread function. In any case, a study of the 2002 behavior of this comet will be presented in a future paper. In the present work, the evidence indicates that the 2001 observations are of a bare cometary nucleus.
Characteristics of C/2001 OG108 (LONEOS)
177
Table 1 Observing parameters of C/2001 OG108 Observing date (2001)
Start time (UT)
End time (UT)
RA (h m)
Dec (◦ ’)
Solar distance
Earth distance
# of obs.
Visible observations (Ondˇrejov Obs. 0.65-m, Lowell Obs. 1.1-m, McDonald Obs. 2.7-m, Univ. of Hawai’i 88-inch) Oct. 4 11:06 12:04 00 19 +22 00 17.00 1.050–1.175 7.17 Oct. 10 18:02 0:36 23 59 +22 24 16.94 1.124–1.574 8.84 Oct. 12 1:19 1:30 23 56 +22 29 16.94 1.596–1.660 9.38 Oct. 12 17:49 2:19 23 54 +22 34 16.93 1.122–2.124 9.68 Oct. 13 17:38 1:14 23 50 +22 38 16.93 1.121–1.644 10.14 Oct. 14 17:50 2:07 23 47 +22 42 16.93 1.121–2.141 10.63 Oct. 15 17:25 0:13 23 44 +22 46 16.94 1.120–1.574 11.13 Oct. 19 17:43 0:46 23 31 +22 57 16.95 1.118–1.728 13.28 Oct. 23 3:17 6:36 23 21 +23 02 16.97 1.021–1.131 15.21 Oct. 24 4:06 6:36 23 18 +23 03 16.98 1.022–1.146 15.80 Nov. 9 18:28 18:28 22 32 +22 46 17.11 1.122–1.222 24.61 Nov. 10 16:50 22:16 22 29 +22 43 17.12 1.121–1.725 25.03 Nov. 12 4:12 4:45 22 25 +22 37 17.13 1.168–1.273 25.68
2.527 2.459 2.446 2.436 2.425 2.414 2.403 2.358 2.321 2.309 2.120 2.109 2.090
1.558 1.505 1.497 1.491 1.485 1.480 1.475 1.459 1.451 1.450 1.476 1.480 1.485
12 14 2 11 7 8 7 5 5 4 1 5 21
Near-infrared observations (NASA Infrared Telescope Facility) Oct. 9 10:01 10:25 00 04 +22 16 Oct. 10 9:21 9:45 00 01 +22 22 9:45 10:01 00 01 +22 22
16.94 16.94 16.94
1.020–1.032 1.003–1.013 1.013–1.016
8.22 8.58 8.58
2.475 2.465 2.464
1.516 1.509 1.508
10 10 6
Mid-infrared observations (Keck I) Oct. 4 11:16 11:48
17.00
1.068–1.132
7.17
2.527
1.558
4
00 19
+22 00
Visual mag.
Airmass range
Phase angle
Note. Solar distance and Earth distance are measured in astronomical units (AU). Phase angle is measured in degrees. RA and Dec are in J2000 coordinates.
Fig. 2. A lightcurve plot for C/2001 OG108 showing its rotation period of 57.2 ± 0.5 h. The up-arrows indicate the times of the NASA IRTF SpeX observations.
2. Observations and reduction
2.1. Visual spectrophotometry
Observations of C/2001 OG108 span three wavelength regimes: visible, near-infrared, and mid-infrared. The details of the observations are listed in Table 1, along with the object’s geometry at the time of the specific observations.
Visual photometric observations were acquired of C/2001 OG108 during October and November 2001 in B, V , R, and I bandpasses (Bessell, 1990). All measurements were calibrated and transformed to the absolute Johnson–Kron– Cousins system using standards from Landolt (1992). Several different telescopes were used to collect these measure-
178
P.A. Abell et al. / Icarus 179 (2005) 174–194
ments, which were subsequently used to generate a complete lightcurve of the comet nucleus (Fig. 2). 2.1.1. Lowell Observatory Observations obtained on October 23 and 24, 2001 Universal Time (UT) were carried out with the Lowell Observatory 1.1-m Hall telescope on Anderson Mesa near Flagstaff, Arizona. CCD photometry was carried out using a SITe back-illuminated 2K × 2K CCD with 24 µm pixels. The frames were binned 2 × 2, resulting in an image scale of 1.1 arcsec per pixel. The usable field of view was 19 arcmin on a side. Whenever a suitable guide star could be found, the telescope was tracked at the object’s nonsidereal rate of motion. The weather was photometric on both nights with seeing estimated to be about 2.7 arcsec. Observations of standard stars were obtained to derive extinction, zero-point, and color transformations, which were in good agreement with other recent observations at the Hall telescope (P. Massey, personal communication). The images were bias-corrected and flat-fielded using twilight flats. Aperture photometry was done with the PHOT task in the IRAF DIGIPHOT package. For all objects, the radius was taken to be 5 pixels and a 5.5 arcsec radius aperture was used. The sky value was found using an annulus from 15–25 pixels radius. Faint background objects were removed by finding the mode of the sky measurements and removing excess values. In all the images C/2001 OG108 appeared to be a point source and showed no evidence of coma. More detailed image analyses were performed to investigate whether a small degree of coma could have been present at the time the observations were obtained. Line profiles of the comet nucleus and comparison stars generated from the data demonstrate similar point-spread functions, indicating that no source extension is observed in the line profile of C/2001 OG108 (Fig. 3). Hence no coma was present during the observations of this comet nucleus. 2.1.2. McDonald Observatory C/2001 OG108 was observed on November 12, 2001 UT from the 2.7-m Harlan J. Smith telescope at the McDonald Observatory located atop Mt. Locke near Fort Davis, Texas. The viewing conditions were good with seeing estimated to be 1.3 arcsec and the skies clear. Measurements of the object were obtained by the Imaging Grism Instrument (IGI) with a 5:1 focal reducer, a Mould R filter, and a TeK 1024 × 1024 CCD. In this configuration the pixel scale is 0.57 arcsec with a usable field of view of approximately 7 arcmin. IRAF photometry routines were used with a 3 arcsec radius aperture to measure the brightness of the object and 10 to 15 reference stars in each image. Relative photometry was used to correct for variations in the extinction of the reference stars from one image to the next. Standard stars were observed over several airmasses, and used to calculate the extinction coefficient and zero point offset in the instrument magnitudes. This information provided a means for calibrat-
Fig. 3. Line profile data of C/2001 OG108 and field stars from images acquired from the Lowell Observatory 1.1-m Hall telescope on October 23 and 24, 2001 UT. The point-spread function (PSF) curve is made by taking a line profile cut through a stack of 8 stars that have been registered and added together. The line profile is taken perpendicular to the direction of motion to minimize the effects from tracking the comet at nonsidereal rates. The solid horizontal line marks the zero point and the dashed lines mark the +1 and −1 sigma levels of the noise in the sky counts associated with the comet image. Note that the pixel values of the comet follow the PSF curve and do not show any evidence of an extended source (i.e., coma) at the time the observations were obtained. There is some deviation from the PSF curve, at distances greater than 4 to 5 arcsec, but this is due to noise in the comet signal which dominates at low count values.
ing the reference stars in each image to an absolute system, which in turn could be used to calibrate the object. To specifically see if C/2001 OG108 exhibited any coma, the radial profile of the stars and the object were compared to each within the same image. In order for this type of comparison to be valid, the images of the stars and the object should not show any signs of trailing. Hence the exposure times were kept short given that the comet nucleus was moving at a significant non-siderial rate. In all the images, neither the object nor the reference stars were trailed by more than half a pixel in each image. No coma of C/2001 OG108 was detected in any of the images during this time. 2.1.3. Ondˇrejov Observatory Observations of C/2001 OG108 were obtained over seven nights from October 10 to 19, 2001 UT and two nights from November 9 to 10, 2001 UT with the Ondˇrejov Observatory 0.65-m telescope. CCD photometry was carried out using a
Characteristics of C/2001 OG108 (LONEOS)
SITe back-illuminated 512 × 512 CCD with 24 µm pixels and a resulting image scale of 2.2 arcsec per pixel. The observational and reduction techniques were essentially identical to those used for earlier near-Earth asteroid photometric work described in Pravec et al. (1996). Color transformations used the measured color indices determined by the participating observatories described in this section and Section 2.2 below. 2.2. Simultaneous visible and mid-infrared observations The October 4, 2001 UT visible dataset was taken with a Tektronix CCD at the University of Hawaii 88-inch telescope on Mauna Kea, Hawai’i. The detector has a pixel scale of 0.22 arcseconds and a field of view of over 7 arcmin. A Johnson V -band and a Cousins R-band filter were used to observe the object, which appeared as a point source in all of the images. Atmospheric conditions were good with the seeing measured to be approximately 0.8 arcsec. The CCD bias level was determined from the median of several zero-exposure images and was subsequently subtracted from each image. All the images of the object also had a flat-field, obtained from the median of several dithered exposures of the blank twilight sky, divided into them. Several calibration stars from Landolt (1992) were observed at various airmasses during the night with stars in the PG+2231, PG+2213, and PG+0918 fields, as well as stars in the vicinity of SA 113-265, SA 92-252, and SA 98-966 being used. The absolute flux calibration, airmass correction, and color correction from the photometry of these stars were simultaneously solved. A total of five exposures in R-band and two in V -band of C/2001 OG108 were obtained; there was negligible change in the photometry from the nucleus’s rotation over the course of the observations. A V -band magnitude of 16.756 ± 0.014 and a V -R color of 0.456 ± 0.016 were derived. The mid-infrared data were taken with the LWS instrument on Keck-I atop Mauna Kea, Hawai’i on October 4, 2001 UT. The detector has a pixel scale of 0.080 arcseconds and a field of view of only 10.2-by-10.2 arcsec. 10%-wide filters centered at 10.7 and 17.9 µm were used to observe the object, and it appeared as a point source in all the images. Seeing was about 0.5 arcsec at both wavelengths. Field flattening was unnecessary since the standard star and the object were observed at the same location on the detector. For absolute flux calibration, the standard star Beta Andromeda was used, which was less than 20◦ from our target and very close in airmass. The star was assumed to have a flux density of 85.9 Jy at 17.9 µm and 235.7 Jy at 10.7 µm; these numbers are derived from the results of Tokunaga (1984). The target was measured in each wavelength twice and from these data, flux densities of 421 ± 20 mJy at 10.7 µm and 742 ± 34 mJy at 17.9 µm were derived.
179
2.3. Near-infrared observations Near-infrared observations of C/2001 OG108 were obtained using the SpeX instrument (Rayner et al., 2003), a medium resolution near-infrared spectrograph, developed by the Institute for Astronomy for the NASA Infrared Telescope Facility (IRTF) on Mauna Kea, Hawai’i. Operating SpeX in the low-resolution (or asteroid) mode and using a 0.8 arcsecond wide slit provides a spectral resolution of ∼93 over the entire ∼0.7 to 2.5 µm wavelength range. Signal-tonoise values of the data are contingent on the brightness of each object observed, the total integration time, and the atmospheric conditions at the summit during observations, but are attainable in excess of 80 to 100 for most objects given good observing conditions. C/2001 OG108 was observed for two nights in October 2001 at an estimated V -Mag = 17 and phase angle of ∼8◦ . Twenty-six 120 second near-infrared spectra were taken of this object between 1.00 and 1.02 airmass on October 9 and 10 UT (Table 1). Summit weather for both nights provided excellent opportunities for observing: the skies were clear with seeing at 0.8 arcseconds or less, the winds calm, and relative humidity levels at values less than 10%. Guide camera images from SpeX showed that these near-infrared spectra were obtained when no coma was detected. Other efforts to study this object two weeks later during its perihelion approach also showed no detection of coma (French, 2002) (Fig. 1). Therefore, the observations of C/2001 OG108 obtained in October were of the surface of the nucleus rather than the dust and gas coma that is typically seen in most comet observations. Observations of a local standard star, SAO 91809, were obtained over a similar airmass range in order to model the atmosphere at Mauna Kea during the observing run. This allows for a more accurate determination of extinction coefficients over the entire spectral interval obtained by the SpeX instrument. These coefficients are contained in starpacks (see Gaffey et al., 2002, for a detailed explanation), which are used to remove the spectral absorption effects of the atmosphere and reduce the noise in the final spectra of the object. Without this type of correction, the spectra will have spurious artifacts due to strong telluric water vapor features, especially at ∼1.4 and ∼1.9 µm, in regions where most of the common materials found in extraterrestrial objects have near-infrared absorptions. Observations were also obtained of a solar analogue star, SAO 93936 (Hyades 64), to correct for any slight differences in the local standard star’s spectrum relative to that of the Sun (Hardorp, 1978). Furthermore, each of the object and standard star spectra were channel-shifted to a registered reference spectrum in order to account for any slight instrument flexure due to the changing orientation of the telescope as it tracked the object/star over the course of the night. All of the C/2001 OG108 , SAO 91809, and SAO 93936 spectra were extracted using the Image Reduction and Analysis Facility (IRAF) program written and distributed
180
P.A. Abell et al. / Icarus 179 (2005) 174–194
Fig. 4. A spectral reflectance plot of C/2001 OG108 with combined visible photometry and near-infrared spectral measurements. The plot is scaled to the determined albedo value of 0.040 at 0.55 µm. The near-infrared data represent a two-night average spectrum of C/2001 OG108 which was taken at the NASA IRTF on October 9 and 10, 2001 UT. The errors are commensurate with the point-to-point scatter in the data.
by the National Optical Astronomy Observatories (NOAO). The raw spectra files were imported into the Spectral Processing Routine (SpecPR) program for processing and analysis where the starpack and channel-shifting corrections are performed (Clark, 1980; Gaffey et al., 2002; Gaffey, 2003). The two-night average near-infrared spectrum of C/2001 OG108 is shown in Fig. 4. A brief examination of the spectrum of C/2001 OG108 demonstrates that this comet nucleus has a nearly linear spectrum with no discrete absorption features, and is generally red in terms of its reflectance at longer wavelengths. Although the overall slope of the spectrum is red, the slope tends to gradually decrease with increasing wavelength. The relative point-to-point scatter of the data shown in Fig. 4 provides a good estimate of the uncertainty associated with these measurements. However, more accurate error analyses using the standard error of the mean indicate that the average uncertainty in the data is roughly 3–5% at wavelengths shorter than 1.8 µm and approximately 8–10% for those points at wavelengths up to 2.3 µm. At wavelengths greater than 2.3 µm, the noise increases to values above 15–20% due to the decreased response of the SpeX detector and the relatively faint (V -Mag = 17, see Table 1) signal from the comet nucleus at the time of the observations.
3. Discussion and analysis Combining the data sets from the three separate wavelength regimes described above gives the opportunity for a more complete study of the comet nucleus. The discussion and analysis that follows provides a detailed characterization of C/2001 OG108 , and is to date the most comprehensive
study of a Halley-type comet nucleus attained via groundbased sensors. 3.1. Lightcurve and pole position All the R-band measurements obtained at the four observatories have been compiled and analyzed for periodicity using the Fourier series fitting procedure developed by Harris et al. (1989) and implemented by Pravec et al. (1996). A unique, single-period solution is obtained of 57.2 ± 0.5 h, with a phase-darkening slope parameter G of −0.01 ± 0.10, and an estimated absolute V magnitude H = 13.05 ± 0.10. The light-travel time corrected lightcurve data (folded with the best fit period, reduced to unit geocentric and heliocentric distances, at a phase angle of 10 degrees, and with the best fit G parameter) is shown in Fig. 2. The rotation phase of 0.0 in Fig. 2 corresponds to JD 2452195.5 = 2001 Oct. 13.0 UT. The data fit well with a single-period Fourier series of the third order and their residuals are consistent with estimated calibration errors. The observed lightcurve is consistent with a principal-axis rotation of the comet nucleus. If there had been any contribution from a nonprincipal rotation (precession) component to C/2001 OG108 ’s observed lightcurve, its amplitude was not greater than a few 0.01 magnitudes. These visible observations indicate that C/2001 OG108 is an elongated object on the basis of its lightcurve with a minimum axial ratio of approximately 1.3 (Fig. 2). The simple sinusoidal and periodic nature of the derived lightcurve suggests that this object shows no evidence of precession, so if any nonprincipal rotation component was present at the time of observation, its effect on the overall amplitude of the lightcurve was negligible (Pravec et al., 2005). A de-
Characteristics of C/2001 OG108 (LONEOS)
rived rotation period of 57.2 ± 0.5 h is somewhat atypical for similarly-sized asteroids (Pravec and Harris, 2000), but is consistent with the measured rotation periods of other Halley-type comets (Samarasinha et al., 2004). Based on the values found in Samarasinha et al. (2004), the rotation period of C/2001 OG108 is the third longest among all of these comets. In addition to deriving the rotation period, it may be possible to constrain the spin axis direction using the fact that the comet was active for only a few months near perihelion. If one assumes that the activity of the comet is driven by seasonal changes on the surface (e.g., coma exists only when an active region on the nucleus’s surface is in sunlight), then the pole’s orientation can be estimated. With the first report of activity coming in early January (Nakamura et al., 2002) and knowing that the activity had ceased by the end of May (Fig. 1), these dates can be adopted to represent the time when the source is in sunlight. Interestingly, this range of dates is centered at approximately the time of perihelion (mid-March) and spans about 75 days before and after. If a pole orientation is to be considered viable, then it must be one in which the source becomes illuminated by January 1 and falls into shadow after late May. Furthermore, the comet’s subsolar latitude should be about the same around January 1 and June 1. Given this configuration, the subsolar latitude must reach its highest value about the time of perihelion, so to first order, one can assume that the pole is pointing toward the Sun at perihelion and the active region is at a high latitude in the illuminated hemisphere. (If the source were near the equator, it would be more likely to produce coma for a longer range of times than was observed.) For this orientation, which corresponds to a possible end state for spin axis migration (Samarasinha and Belton, 1995), the pole is pointed to an ecliptic latitude of −62 degrees and a longitude of 352 degrees (or in equatorial coordinates, RA = 32 degrees, Dec = −57 degrees). Note that this result is only approximate and tests indicate that deviations of up to 40 degrees in some directions can still produce viable configurations for reproducing the observed coma. Furthermore the short-lived coma could simply be the result of an outburst event, which would make our fundamental assumptions invalid. If the above solution is indeed the true pole position, then the sub-Earth latitude during the time the lightcurve measurements were taken (about 150 days prior to perihelion) was located within a few degrees of the equator. Hence, the observed lightcurve amplitude is the maximum possible and the axial ratio of C/2001 OG108 is close to 1.3 (Fig. 2). However, the uncertainties in the pole position means that the observed amplitude could represent the rotation of a body that is somewhat foreshortened. Previous observations of lightcurves have been used to constrain the axial ratios of comets, even when the pole position is not precisely known, as demonstrated by the following equation
taken from (Fernández et al., 2000) (a/c)2 + tan2 l , δm = 1.25 log (b/c)2 + tan2 l
181
(1)
where a/c and b/c are the axial ratios of the nucleus and l is the sub-Earth latitude on the comet’s surface. Therefore a difference of 40 degrees in sub-Earth latitude would give an upper limit to the axial ratio of approximately 1.5 for C/2001 OG108 . Either axial ratio value discussed above is consistent with observations obtained from other Oort cloud comets (e.g., 1P/Halley). 3.2. Color indices and phase slope parameter Visual photometric observations of C/2001 OG108 in R-filter combined with several additional points obtained in B, V , and I -filters (Johnson–Cousins system) give color indices of: B-V = 0.76 ± 0.03; V -R = 0.46 ± 0.02; V -I = 0.90 ± 0.02 and a phase slope parameter G of −0.01 ± 0.1. The values of V -R collected in this study are almost identical to those determined for extinct comets (0.44 ± 0.02) and comet nuclei (0.45 ± 0.02) (Jewitt, 2002). In addition, the color indices of C/2001 OG108 are similar to those determined for Trojan D-type asteroids and Damocloids (Table 2), and are not unlike the average values determined for Damocloids by Jewitt (2005). These results suggest that there may be some similarities in the surface materials of all of these objects as has been noted by many others (Hartmann et al., 1987; A’Hearn, 1988; Fitzsimmons et al., 1994; Di Martino et al., 1998; Davies et al., 1998, 2001; De Sanctis et al., 2000; Hicks et al., 2000; Harris et al., 2001; Licandro et al., 2002, 2003). Some centaurs show a certain degree of affinity in terms of their color indices to C/2001 OG108 , whereas others, such as (5145) Pholus, clearly are not similar in terms of composition to this comet based on this criterion (Bauer et al., 2003). Kuiper belt objects generally do not exhibit similar color indices to this object and are not suitable analogues for the composition of this Halley-type comet (Table 2) (Jewitt and Luu, 2001). The determined value of the phase slope parameter G of −0.01 ± 0.1 was calculated by fitting the visual photometric data to an H –G phase relationship. A phase slope parameter G was fitted simultaneously with the measured period of the comet nucleus and a range of values from −0.2 to 0.5 was grid-searched in increments of 0.01 while the period was solved for in each increment. The best fit solution was found to be G = −0.01 and a formal estimate of the error in G based on a chi2 statistic of the fit gave a 1 sigma uncertainty of 0.02. A more realistic uncertainty of G, taking into account the possible systematic errors due to absolute calibration errors in the datasets taken at the different observatories, as well as possible minor aspect changes during the observational interval, is likely a few times greater. Therefore, 0.1 is a more realistic uncertainty of the derived
182
P.A. Abell et al. / Icarus 179 (2005) 174–194
Table 2 Color indices and slope parameters of outer Solar System objects Object name
B-V
V -R
V -I
G
Notes
Reference
C/2001 OG108
0.76 ± 0.03
0.46 ± 0.02
0.90 ± 0.02
−0.01 ± 0.1
Comet
This work
(624) Hektor
0.77
0.44
0.950
–
Trojan (D Type)
a,b
1996 PW 1998WU24
– 0.78 ± 0.034
0.56 ± 0.04 0.53 ± 0.037
1.03 ± 0.06 0.99 ± 0.035
Low (0.05) –
Damocloid Damocloid
c
1P/Halley 2P/Encke 28P/Neujmin 1 49P/Arend-Rigaux 1
0.73 ± 0.03 – 0.86 ± 0.04 –
0.44 ± 0.03 0.46 ± 0.02 0.41 ± 0.03 0.47 ± 0.01
0.92 ± 0.07 – 0.99 ± 0.04 0.98 ± 0.02
– −0.252 +0.41 ± 0.08 –
Comet Comet Comet Comet
e
(5145) Pholus 1992 AD (10199) Chariklo 1997 CU26 (10370) Hylonome 1995 DW2 (49036) Pelion 1998 QM107 (52872) Okyrhoe 1998 SG35
– – – – –
0.78 ± 0.05 0.49 ± 0.02 0.50 ± 0.07 0.63 ± 0.12 0.42 ± 0.08
1.59 ± 0.05 1.00 ± 0.02 1.02 ± 0.07 1.27 ± 0.11 0.88 ± 0.06
+0.16 ± 0.02 +0.08 ± 0.05 +0.18 ± 0.10 −0.11 ± 0.08 +0.12 ± 0.07
Centaur Centaur Centaur Centaur Centaur
k
(26375) 1999 DE9 (38628) Huya 2000 EB173 (79360) 1997 CS29
0.94 ± 0.03 0.93 ± 0.04 1.16 ± 0.06
0.57 ± 0.03 0.65 ± 0.03 0.61 ± 0.05
1.13 ± 0.03 1.24 ± 0.03 1.27 ± 0.05
−0.34 ± 0.05 +0.10 ± 0.03 −1.42 ± 0.30
Kuiper belt object Kuiper belt object Kuiper belt object
l,m
a b c d e f g h i j k l m
d
f,g h,i j
k k k k
l,m l,m
Dunlap and Gehrels (1969). Degewij and Van Houten (1979). Davies et al. (1998). Davies et al. (2001). Thomas and Keller (1989) deduced by Davies et al. (1998) from reflectivity gradients. Fernández et al. (2000). Jewitt (2002). Campins et al. (1987) deduced by Davies et al. (1998) from reflectivity gradients. Delahodde et al. (2001). Luu (1993) deduced by Davies et al. (1998) from reflectivity gradients. Bauer et al. (2003). Jewitt and Luu (2001). Sheppard and Jewitt (2003).
G parameter for C/2001 OG108 . Such a value of G is indicative of an object with a low albedo, and is consistent with the albedo value derived from the simultaneous visible and mid-infrared data (see Section 3.3). This value is also consistent with those derived from other primitive, low albedo objects such as comets, centaurs, and Kuiper belt objects (Table 2) (Fernández et al., 2000; Sheppard and Jewitt, 2003; Buratti et al., 2004). 3.3. Albedo and size The simultaneous visible and mid-infrared observations allow for a relatively robust determination of the object’s radius and albedo. This technique has been used for over 30 years and basically involves solving two equations, one for the visible flux and one for the mid-infrared flux, with two unknowns (Allen, 1970, 1971). The method is described by Lebofsky and Spencer (1989) in an asteroid context and by Lamy et al. (2004) in a comet context, and it basically rests on knowing the object’s surface temperature map at the time of observation. This map is derived from a thermal model, and in the vast majority of cases the slow-rotator thermal model of Lebofsky et al. (1986), with an update by Harris (1998), gives an adequate description of the temperature [see
the review by Harris and Lagerros (2002) for details]. The slow-rotator model name is somewhat misleading since for most objects the thermal inertia is more important than the actual rotation rate. In any case, the largest uncertainty in the model comes from the so-called beaming parameter (η), which accounts for the fact that the surface of a real object has topography and will not radiate the same way a smooth sphere or ellipsoid will. The model has a few other parameters, all of which are either better known or less influential on the results; a detailed discussion of them is given by Lamy et al. (2004). As shown by Harris et al. (1998) and Delbo et al. (2003), the actual value of η can vary from object to object, and may also be somewhat dependent on the phase angle. Thus it is difficult to confidently assume a single value for η without any other prior knowledge of the object in question; instead a more conservative approach would be to use a range of values for η. For example, if η = 1.0 is applied to the observations of C/2001 OG108 , a value that seems appropriate for Comet 2P/Encke (Fernández et al., 2004), then the comet has a V -band geometric albedo pV = 0.030 ± 0.005, an R-band geometric albedo pR = 0.032 ± 0.005, and an effective radius RN = 8.8 ± 0.2 km. However, Fernández et al. (2005) suggest that a very low value of η of approximately 0.52
Characteristics of C/2001 OG108 (LONEOS)
to 0.75 may be warranted for C/2001 OG108 instead, which would produce a pV = 0.050 ± 0.008, a pR = 0.054 ± 0.008, and an RN = 6.8 ± 0.5 km. Such low values for η are somewhat unusual, and given the difficulties of observing in the Q-band, a slight error in the mid-IR measurements could affect the assumed range of η used in the thermophysical modeling process, and thus produce a larger albedo value for the comet nucleus. Because of the uncertainty associated in determining the correct value for the beaming parameter of C/2001 OG108 , the following mid-range values with appropriate errors will be used to characterize this object: pV = 0.040 ± 0.010, pR = 0.043 ± 0.010, and RN = 7.6 ± 1.0 km. This value of the radius applies to the actual time of observations, which were done at a rotation phase of 0.43 (Fig. 2). Given the photometric range of the light curve, this effective radius can be converted to the effective radius at the maximum of the light curve. For a prolate ellipsoid with semi-major axes a > b = c, the cross section at the light curve maximum should be proportional to a ∗ b. Thus from the mid-IR data, the individual semi-major axes can be calculated as follows: the light curve maxima occur at phases 0.23 and 0.73, and the object is about 0.22 mag brighter than at a phase of 0.43 (Fig. 2). Since the mid-IR flux is proportional to the square of the radius, if C/2001 OG108 was observed in the mid-IR at the maximum of the light curve, a cross section of a ∗ b = (7.6 ± 1.0 km)2 ∗ 10(0.4∗0.22) = (8.4 ± 1.0 km)2 would have been obtained. For axial ratio a/b = 1.3, this leads to a = 9.6 ± 1.0 km and b = 7.4 ± 1.0 km. Note that an important uncertainty in this radius estimate, aside from the beaming parameter discussed above, is the viewing geometry of the nucleus at the time the observations were obtained. If the comet nucleus was not observed at an equatorial viewing geometry, then the nucleus’s true axial ratio could be larger than 1.3 (see Section 3.1). The albedo measurement of C/2001 OG108 agrees well with previous values obtained from asteroids that have been identified as good candidates for possible extinct comet nuclei (Harris et al., 2001; Fernández et al., 2005). Six of these objects are also Damocloids and their radii and albedos are shown for comparison in Table 3. Note that the albedo of C/2001 OG108 falls within the range of the error bars
183
for all but one of the objects listed, and is quite similar in terms of size to Damocloids (65407) 2002 RP120 , 1999 LE31 , and 2000 DG8 . As discussed above, these types of asteroids may be possibly related to the Halley-type comets on the basis of their orbital parameters (see Section 1). The data from Table 3 suggests that there may not only be a similarity in orbital parameters with these types of objects, but that some Damocloid asteroids and Halley-type comets may also have similar physical properties in terms of size and albedo. In addition, the albedo value determined for C/2001 OG108 is also consistent with those derived for the nuclei of active comets (Campins and Fernández, 2002; Lamy et al., 2004). There seems to be as yet no obvious difference in the albedo distribution of the nuclei of active comets and that of Damocloids/extinct comet candidates observed to date. 3.4. Possible meteorite affinities The determined albedo value of 0.040 ±0.010 and visible measurements obtained are applied to the best average near-infrared spectrum of C/2001 OG108 (Fig. 4) so that a comparison of its spectral reflectance to those of other extraterrestrial materials with similar albedo values can be performed. Before the albedo adjustment was made, all the visible data were scaled to the near-infrared data at approximately 0.79 microns, which is the center wavelength for the Cousins I-filter. This was the only wavelength overlap between the two spectral data sets obtained for C/2001 OG108 . Extraterrestrial objects that have similar albedos to this comet, and that have reliable visible and near-infrared spectral measurements, are the carbonaceous chondrite meteorites (Gaffey, 1976). Apart from a few exceptions, these meteorites are easily distinguished from most other chondritic meteorite groups due to their dark visual appearance in hand sample. A more detailed examination of these meteorites reveals the presence of a fine-grained (5 µm), opaque matrix, which may contain 1 to 5 weight percent carbon (Brearley and Jones, 1998) and nickel-iron sulfides (M. Zolensky, personal communication). The combination of the fine-grained matrix, carbon, and/or nickel-iron sulfides produces the low albedos observed in these meteorites
Table 3 Radii and albedos of Damocloids and C/2001 OG108 Object C/2001 OG108 (20461) Dioretsa 1999 LD31 (65407) 2002 RP120 1999 LE31 2000 DG8 2000 HE46 2003 WN188
Effective radius (km)
Geometric albedo
Reference
7.6 ± 1.0
0.040 ± 0.010
This work
14.0 ± 3.0 7.3 ± 1.4 8.4 ± 2.1 7.8 ± 1.3 3.2 ± 0.6 5.0 ± 1.1
0.030 ± 0.010 0.090 ± 0.036 0.052 ± 0.026 0.049 ± 0.017 0.041 ± 0.016 0.046 ± 0.021
a
Note. Errors listed are the maximum range of values that yield acceptable fits to a standard thermophysical model. a Harris et al. (2001). b Fernández et al. (2005).
b b b b b
184
P.A. Abell et al. / Icarus 179 (2005) 174–194
(a)
(b) Fig. 5. (a) A comparison of C/2001 OG108 ’s spectral reflectance with carbonaceous meteorites. All meteorite data, except those of Tagish Lake, are from Gaffey (1976). Tagish Lake data are from Hiroi et al. (2001). (b) The same comparison of C/2001 OG108 ’s spectral reflectance with carbonaceous meteorites as shown in (a), but with the vertical scale expanded for clarity.
(Gaffey, 1976; Clark and Lucey, 1984; Buseck and Hua, 1993; M. Zolensky, personal communication). A plot of the spectral reflectance of C/2001 OG108 versus several carbonaceous chondrite meteorites is shown in Fig. 5a. It should be noted that the data obtained of C/2001 OG108 is of a whole disk integrated average, which could mask any albedo and/or spectral variations due to differences in particle size, mineralogy, or abundance, whereas the meteorite spectra are of well-characterized powders measured under known laboratory conditions. Upon comparing the spectra of select meteorites from this group to the data of C/2001 OG108 , it is immediately
clear that no meteorite sample currently in the terrestrial collections matches the spectrum of this comet nucleus. All of the characterized meteorite types listed in Fig. 5a, with the exception of Tagish Lake, have albedos that fall above the value of 0.040 ± 0.010 that has been determined for C/2001 OG108 . In addition, all of the carbonaceous meteorites show evidence of absorption features in their spectra, whereas the comet nucleus does not appear to have any such features at the detection limit of these data (Fig. 5b). Those meteorites, belonging to the CV3 class (e.g., Allende and Mokoia), the CM2 class (e.g., Murchison and Murray), and the CI1 class (e.g., Orgueil), have evident features due to
Characteristics of C/2001 OG108 (LONEOS)
the presence of olivine, pyroxene, or phyllosilicate minerals, which produce features in the 1 and 2 µm regions of their near-infrared spectra and strong absorption edges in their UV-visible spectra (Gaffey, 1976; Gaffey et al., 1993; Calvin and King, 1997). Tagish Lake, on the other hand, does not appear to have features as obvious as those seen in the other meteorites shown in Fig. 5b, but has a broad, weak feature located near 1 µm seen in the spectrum measured by Hiroi et al. (2001). Although none of the meteorites shown in Fig. 5a match the spectral response of C/2001 OG108 , Tagish Lake seems to have a spectrum that is more similar to the comet nucleus than any of the other samples. It has the same slight increase in slope in the visible portion of the spectrum as the comet nucleus, which is unlike the steeply rising spectra seen in the other carbonaceous meteorites over this same wavelength region, and has a similar response in the near-infrared to the spectrum of C/2001 OG108 . However, the presence of the broad, weak feature located near 1 µm seen in the Tagish Lake spectra (Fig. 5b), and the observed hydrated phases of this meteorite (Brown et al., 2000), may preclude any mineralogical association with this comet nucleus. It is important to note that C/2001 OG108 is not a plausible parent body for the Tagish Lake meteorite, even though some similarities exist between their spectra, due to the significant differences in their orbital parameters. Brown et al. (2000) report that the derived orbital elements place Tagish Lake and its parent body in an orbit similar to that of an Apollo asteroid, with a semi-major axis well inside the main asteroid belt. C/2001 OG108 has a much larger semimajor axis and its orbital characteristics are similar to the Halley-type comets (Asher et al., 1994; Bailey and Emel’yanenko, 1996). In addition, the initial entry velocity of Tagish Lake is estimated to be 15.8 ± 0.6 km/s (Brown et al., 2000),
185
whereas any entry velocity of a similar sized fragment of C/2001 OG108 would be approximately 48 km/s and would almost certainly cause the fragment to vaporize in the upper atmosphere (Flynn, 1989). 3.5. Possible asteroid affinities Other extraterrestrial objects that have similar low albedo values with relatively featureless spectra are asteroids of the F, C, P, T, and D taxonomic classes (Tholen, 1984; Tholen and Barucci, 1989). These asteroids have albedos typically ranging from 3 to 8% (Tedesco et al., 1992, 2002) and may be good analogues for a comparison to the spectral reflectance of C/2001 OG108 . F-type asteroids generally have flat, reddish spectra at wavelengths longwards of 0.4 µm, but differ from the C-type asteroids in the visible portion of their spectra even though the C-types have a similar flat to slightly red reflectance at longer wavelengths. The other three remaining groups have a much more reddish spectral response in the near-infrared regions of their spectra than either the F- or C-types (Gaffey et al., 1993). P- and T-type asteroids both have significant red slopes to their spectra, but the T-types tend to flatten out with increasing wavelength (Tholen and Barucci, 1989; Britt et al., 1992; Gaffey et al., 1993). Of the five listed classes, the reddest spectrum belongs to asteroids in the D taxonomic class. These asteroids have spectra that are slightly red in the visible wavelengths up to 0.55 µm, but extremely red in visible and near-infrared wavelengths longer than 0.55 µm (Bell et al., 1988; Tholen and Barucci, 1989; Vilas and Gaffey, 1989; Gaffey et al., 1993). Several asteroid spectra belonging to the F, C, P, T, and D taxonomic classes have been plotted in comparison to the spectrum of this comet nucleus and are shown
Fig. 6. Representative asteroid spectra for each of the low albedo taxonomic classes is plotted here and compared with the spectrum of C/2001 OG108 . All spectra are normalized to unity at 0.55 µm and then offset for clarity. Asteroid spectra are taken from Zellner et al. (1985) and Bell et al. (1988).
186
P.A. Abell et al. / Icarus 179 (2005) 174–194
in Fig. 6. The spectra of F- and C-type asteroids and this comet nucleus are clearly quite dissimilar as the slopes of these asteroid spectra are generally much flatter than that of C/2001 OG108 . Spectra belonging to members of the T-type asteroids match the steep visible response of C/2001 OG108 , but tend to flatten out their spectral slopes with increasing wavelength. P-type asteroids have similar slopes in the near-infrared portions of their spectra to the comet nucleus, but do not match the visible portion of its spectrum, which has a much steeper spectral response. Of all the asteroid taxonomic types, the reflectance spectrum of C/2001 OG108 appears to most closely resemble that of a D-type asteroid (Fig. 6). The D-type spectra presented here match both the visible and near-infrared portions of the comet’s nucleus, unlike the T- and P-type spectra, which only match either the visible or near-infrared portions of the spectrum, respectively. The observation that this comet nucleus has a reflectance spectrum similar to those of D-type asteroids is consistent with previous observations of other cometary nuclei and extinct comet candidates. Several investigators have suggested similarities between the characteristics of D-type asteroids and comet nuclei in terms of their spectral measurements and albedos (e.g., Hartmann et al., 1987; A’Hearn, 1988; Fitzsimmons et al., 1994; Di Martino et al., 1998; Davies et al., 1998, 2001; De Sanctis et al., 2000; Hicks et al., 2000; Harris et al., 2001; Licandro et al., 2002, 2003). D-type asteroids typically have low albedos (0.05) and are similar to values seen for most comet nuclei (0.03) (A’Hearn, 1988). Although some comets have been reported to have slightly higher albedos (0.04 to 0.06), they are still within range of the albedos seen for D-types given the reported uncertainty of 0.01 to 0.03 in the data for those observations (Campins and Fernández, 2002). Asteroids of the T taxonomic class have a higher range of albedos (0.10), whereas the P-types have a similar albedo range (0.04) to those seen for the D-types and cometary nuclei (Gaffey et al., 1993). In addition to the spectral and albedo similarities between C/2001 OG108 and these asteroids, it also seems that D-type asteroids have a plausible range of compositions that is suited to the current understanding of what comet nuclei may be composed of in terms of organic content, anhydrous silicates, and water ice (A’Hearn et al., 1995; Joswiak et al., 2001; Soderblom et al., 2002). However, Ptype asteroids may also fit these compositional criteria and should not be excluded as potential comet analogues on this basis alone as future observations of comets may show some similarities to these objects. It is important to note that not all spectra of comet nuclei may resemble those of D-type asteroids, given the limitations of the taxonomic classification schemes used, but the similarity of the spectra shown here does suggest the possibility that C/2001 OG108 , a Halleytype comet, has similar characteristics to this taxonomic class of asteroids.
3.6. Spectral comparison with interplanetary dust particles Detailed examination of the average spectrum of C/2001 OG108 (Fig. 4) reveals that there is no indication of the presence of weak 0.8 to 1.0 µm features, at a 5% detection limit, such as are seen in the spectra of carbonaceous chondrites and many low-albedo asteroids (Vilas and Gaffey, 1989; Vilas et al., 1994; Calvin and King, 1997). This would be consistent with the presence of anhydrous rather than hydrous silicates, similar to the results of measurements of dust from Comet 1P/Halley made by the Giotto spacecraft and results from observations made by the Deep Space 1 probe of Comet 19P/Borrelly (Brownlee, 1988; Soderblom et al., 2002). It is important to note that although the possibility exists for C/2001 OG108 to have near-infrared spectral features below the detection limit for this data set, any features are likely to be due to anhydrous silicates, and may not be readily detected via ground-based observations due to the low albedo of this comet nucleus. No known recognized macroscopic sample of a comet nucleus exists in the terrestrial collections. However, small 5 to 15 µm particles, referred to as interplanetary dust particles (IDPs), have been identified as possibly having cometary origins on the basis of their heating profiles from high velocity (V > 18 km/s) atmospheric entry (Flynn, 1989; Jackson and Zook, 1992; Nier and Schlutter, 1993; Brownlee et al., 1994, 1995; Joswiak et al., 2001). To date, of the 13 IDPs that have been identified as having an entry velocity consistent with a cometary origin, 12 have anhydrous silicate mineralogies with 11 members of this group having significant amounts of amorphous carbon (Joswiak et al., 2001). The sole remaining high velocity IDP (V = 20.4 km/s) is reported to be phyllosilicate-rich, and investigators Joswiak et al. (2001) suggest that it may either be a unique type of cometary IDP or an asteroidal IDP whose parent body was perturbed into a high velocity orbit relative to Earth prior to the particle’s atmospheric entry. Most anhydrous IDPs that have been measured through reflectance spectroscopy appear to have low albedos (
