Imaging and photometry of comet C/1999 s4 (LINEAR) before perihelion and after breakup

Science

Volumn 292, Issue 5520, 2001, Pages 1348-1353

  Farnham, Tony L ^a   Schleicher, David G ^b   Woodney, Laura M ^b   Birch, Peter V ^c   Eberhardy, Clara A ^d   Levy, Lorenza ^e  

^a University of Texas at Austin   (United States)

^b LOWELL OBSERVATORY   (United States)

^c PERTH OBSERVATORY   (Australia)

^d UNIVERSITY OF WASHINGTON   (United States)

^e NORTHERN ARIZONA UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

CARBON; DATA ACQUISITION; IMAGE ANALYSIS; PHOTOMETRY; SPACE TELESCOPES;

PERIHELION;

SOLAR SYSTEM;

CARBON;

COMET; IMAGING METHOD; PHOTOMETER;

ARTICLE; ASTRONOMY; IMAGE ANALYSIS; MEASUREMENT; MOLECULAR MODEL; PHOTOMETRY; PRIORITY JOURNAL; ROTATION;

EID: 0035907107     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.1058886     Document Type: Article

References (42)

1. Narrowband photometry was performed using the HB comet filters (39) to isolate flux from gas species and the underlying dust continuum. For CCD imaging, we typically used broadband V and R filters (554/85 nm and 650/120 nm, respectively), although when the comet was bright enough we also used the HB narrowband filters to investigate any structure present in the gas emission.
2. Gas production rates were computed using the standard Haser model and our nominal model scale lengths. A(θ)fρ is a proxy for dust production, where A(θ) is the albedo at the particular phase angle of the solar illumination at the time of the observation, f is the filling factor of the grains within the field of view of the photometer entrance aperture, and p is the projected radius of the aperture. The quantity A(θ)fρ is aperture-independent for dust having a canonical 1/p spatial distribution, and is independent of wavelength for grains that are gray in color.
3. 3, and λ345 nm) were not always observed. Additionally, the relatively high dust-to-gas ratio seen early in the apparition, coupled with the low signal-to-noise ratio, resulted in low or negative gas fluxes after subtraction of the underlying continuum.
4. D. Schleicher, C. Eberhardy, IAU Circular 7455 (2000).
5. H. A. Weaver et al., Science 292, 1329 (2001).
6. J. T. T. Mäkinen, J.-L. Bertaux, M. R. Combi, I. Quémerais, Science 292, 1326 (2001).
7. The CN production rate curve was transposed to each of the other species by applying a simple scaling factor to the production rates based on the relative abundances measured in June and July.
8. A dynamically new comet is one having an orbit that is statistically likely to occur only upon the comet's initial return to the inner solar system from the outer Oort Cloud. Subsequent passages will be in modified orbits because of perturbations by the giant planets.
9. F. L. Whipple, Moon Planets 18, 343 (1978).
10. M. F. A'Hearn, R. L. Millis, D. G. Schleicher, D. J. Osip, P. V. Birch, Icarus 118, 223 (1995).
11. D. C. Jewitt, K. J. Meech, Astrophys. J. 310, 937 (1986).
12. H < 3 AU, we believe this is a valid assumption.
13. 2 by 18 July. These areas are systematically lower than our values because of a combination of somewhat lower derived water production rates and differences between the water vaporization models.
14. Seasonal variations could also produce a decrease in activity; however, in this case, the effect goes in the wrong direction. Imaging data discussed below indicate that the Sun moved from near the equator toward the pole during this time, which would be expected to produce an increase in the total activity (10).
15. A. L. Cochran, W. D. Cochran, Bull. Am. Aston. Soc. 32, 1070 (2000).
16. H. F. Levison, M. J. Duncan, Icarus 127, 13 (1997).
17. D. Bockelée-Morvan et al., Science 292, 1339 (2001).
18. M. J. Mumma et al., Science 292, 1334 (2001).
19. When the comet was outside of solar conjunction, we have approximately monthly coverage. During the months of January and July, however, we have extensive observations.
20. The length of the plume is 5000 km as projected onto the plane of the sky. It could be longer if the jet points toward or away from Earth. This plume is discernible even in unprocessed images, so we are confident that it is not an artifact of the image enhancement.
21. We searched, unsuccessfully, for this fragment in our data. Estimates of the separation speed (5) indicate that even on 9 July, the fragment would be too close to the nucleus to be resolved in our images.
22. G. P. Tozzi, J. Licandro, in preparation.
23. If the spin axis lies near the plane of the sky, then throughout most of the rotation, the jet's emission is directed roughly toward or away from Earth, and no structure can be discerned. At two points during the rotation, however, the vent lies near the plane of the sky, and material emitted from the jet at these times is seen side-on. As this dust moves out from the nucleus and gets pushed away from the Sun by radiation pressure, it is seen as the wings.
24. For each date, the rotation axis is located in a plane defined by the line of sight and the projection of the rotation axis onto the sky (e.g., the line of symmetry). The intersection of planes from different dates is then used to determine the pole orientation in three dimensions.
25. The SWAN instrument on SOHO monitored the entire neutral hydrogen coma of C/LINEAR from late May to mid-August (6) and recorded a number of outbursts. The large size of the hydrogen coma introduces a significant phase lag (1 to 2 days) and dilution in amplitude of the observed peak in hydrogen abundances as compared to the instantaneous activity at the nucleus caused by an outburst.
26. M. Kidger, IAU Circular 7467 (2000).
27. M. L. Finson, R. F. Probstein, Astrophys. J. 154, 353 (1968).
28. The F-P technique uses the "sorting" effects of radiation pressure on different-sized grains to model the surface brightness of the dust tail. For any given date, it is possible to compute the location, relative to the nucleus, of a given particle radius a emitted at an earlier time τ. In accord with this concept, a and τ are used to compute the positions of syndynes (the locus of particles of a single size emitted over a range of times) and synchrones (the locus of different-sized particles all emitted at a specific time), which map the distribution of grains in the tail. The expected surface brightness is computed by adding the contributions from all different dust sizes emitted over all times of interest The particle size limits in the model are set by the information content of the images, and the amount of light reflected by each combination of a and τ is determined by the model parameters outlined in the text.
29. Outward expansion of the dust after it leaves the nucleus causes the surface brightness contribution from one set of (a, τ) values to overlap with that from other sets. Thus, a small change in one parameter can be offset by a change in another parameter, allowing multiple solutions to reproduce the tail profile in an image. For comets with widely spaced syndynes and synchrones (e.g., Fig. 3), this cross-parameter variation can be reduced to 5% or so, which is small enough to have no effect on the conclusions presented here. In addition, modeling a sequence of images introduces temporal constraints that reduce the cross-parameter variations even more.
30. Photometry after the breakup (Fig. 1) shows that the gas production dropped by at least a factor of 5, whereas the amount of dust was reduced by a factor of 10. Also, HST images (5) show that by 5 August, the large fragments were emitting a negligible amount of dust compared to that already in the tail.
31. Complications introduced by the breakup include numerous fragments emitting dust rather than a single source, a changing PSD resulting from fragmenting grains, and light from ion and gas species that cannot be separated from the light reflected by dust. Although we cannot tightly constrain the effects of these problems, we believe that they will not affect the parameters at a level of more than a few percent. In addition, their influence is most severe immediately after the disruption and diminishes with time. For example, the numerous large fragments are all very close together relative to the tail dimensions, so they mimic a single nucleus. By the time they have moved apart, their activity levels have also dropped significantly, as seen in the HST images (5). One problem that gets worse with time is that there is no well-defined nucleus to use as a reference point for the dust motions. We minimize this problem by modeling several images in a sequence, because constraints from the first image, where the nucleus is defined, will propagate through successive images.
32. 4 times as much light).
33. W. Waniak, Icarus 100, 154 (1992).
34. M. Fulle, Astron. Astrophys. 276, 582 (1993).
35. A. Kammerer, Int. Comet Q. 22, 71 (2000).
36. -3 for all computations in this report, with the acknowledgment that this could be off by a factor of ∼2 in either direction.
37. An estimate of the hidden mass can be found by extrapolating the dust PSD to particles up to 10 m in radius PSD indices as steep as -3.5 can produce enough mass to account for a 0.44-km nucleus. Therefore, the hidden mass computed for the PSD index (-2.4) found for 30- to 300-μm grains in our model is sufficient to produce a parent body even larger than 0.44 km.
38. In our analysis, we assume that all of the ice in the nucleus was exhausted during the breakup. The fact that very little water production was measured after the breakup supports this assumption.
39. T. L. Farnham, D. G. Schleicher, M. F. A'Hearn, Icarus 180, 147 (2000).
40. J. J. Cowan, M. F. A'Hearn, Moon Planets 21, 155 (1979).
41. Application of this processing technique was slightly different for the two dates. The July images were rotated directly around the nucleus, whereas the June image was unwrapped in (r,θ) space, shifted in the 6 axis, and then rewrapped into the (x,y) image. The only difference between these applications is that the innermost two or three pixels of the June image are more sharply enhanced.
42. We thank the authors of the other reports in this special C/LINEAR edition, in particular H. Weaver, M. Combi, C. Lisse, and D. Bockelée-Morvan, for discussions that were useful in incorporating the different data sets into a cohesive picture. Supported by NASA grants NAG5-4384, NAG5-7947, and NAG5-9009 and by NSF grant NSF 9988007.