1
T.L. Farnham
a
, , D.D. Wellnitz
a
, D.L. Hampton
b
, J.-Y. Li
a
, J.M. Sunshine
a
, O. Groussin
a
,
L.A. McFadden
a
, C.J. Crockett
a
, M.F. AHearn
a
, M.J.S. Belton
c
, P. Schultz
d
, C.M. Lisse
e
a
Department of Astronomy, University of Maryland, College Park, MD 20742, USA
b
Ball Aerospace and Technology Corp., 1600 Commerce Street, Boulder, CO 80301, USA
c
Belton Space Exploration Initiatives, 430 S. Randolph Way, Tucson, AZ 85716, USA
d
Department of Geological Sciences, Brown University, Providence, RI 02412, USA
e
Applied Physics Laboratory, Johns Hopkins University, Laurel, MD 20723, USA
Received 15 June 2006; revised 25 October 2006
Available online 16 January 2007
Abstract
We present an overview of the dust coma observations of Comet Tempel 1 that were obtained during the approach and encounter phases of the
Deep Impact mission. We use these observations to set constraints on the pre-impact activity of the comet and discuss some preliminary results.
The temporal and spatial changes that were observed during approach reveal three distinct jets rotating with a 1.7-day periodicity. The brightest jet
produces an arcuate feature that expands outward with a projected velocity of about 12 m s
1
, suggesting that the ambient dust coma is dominated
by millimeter-sized dust grains. As the spatial resolution improves, more jets and fans are revealed. We use stereo pairs of high-resolution images
to put some crude constraints on the source locations of some of the brightest features. We also present a number of interesting coma features that
were observed, including surface jets detected at the limb of the nucleus when the exposed ice patches are passing over the horizon, and features
that appear to be jets emanating from unilluminated sources near the negative pole. We also provide a list of 10 outbursts of various sizes that
were observed in the near-continuous monitoring during the approach phase.
�
2007 Elsevier Inc. All rights reserved.
Keywords: Comet Tempel-1; Comets; Rotational dynamics
1. Introduction
Recent studies of comets suggest that most comet nuclei
are essentially inactive, with fractional active areas typically
measured to be a few percent (
AHearn et al., 1995; Keller
et al., 2004
). Observations of jets, arcs and other features
prompted suggestions that signicant amounts of the activity
in many comets were produced from isolated active areas or
vents (e.g.,
Sekanina, 1991; Keller et al., 2004
, and references
therein). Indeed, as ground-based observations have improved
over the years, more and more comets have been observed
to have coma features. Even comets that appear featureless at
large distances reveal structure as they get close to the Earth,
suggesting that, with enough resolution, features could be ob-
*
Corresponding author. Fax: +1 (301) 405 3538.
E-mail address:
farnham@astro.umd.edu
(T.L. Farnham).
served in the coma of any comet. In support of this conjec-
ture, each of the four comets imaged by spacecraft (1P/Halley,
19P/Borrelly, 81P/Wild 2, and 9P/Tempel 1) exhibited numer-
ous well-dened, and even collimated, jets in their inner comae
(e.g.,
Sagdeev et al., 1985
;
AHearn et al., 1986, 2005
;
Sekanina
and Larson, 1986; Soderblom et al., 2002; Brownlee et al.,
2004
), including small-scale features that were not resolved
in ground-based observations. The existence of these features
in so many different objects suggests that there is a common
mantling process (or other process that restricts local activity)
that acts to insulate most of the comets surface from sublima-
tion. This mantle forms a protective covering that seals off the
volatile material underneath. Only at breaks or thin spots in the
mantle does activity persist to produce the coma. There are ex-
ceptions, including Comet 46P/Wirtanen, where large fractions
of the surface appear to be active (
Farnham and Schleicher,
1998; Groussin and Lamy, 2003
), which indicates that heavy
mantling may not be a completely universal phenomenon. It is
0019-1035/$ see front matter
� 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.icarus.2006.10.036 Dust morphology in Comet 9P/Tempel 1
27
possible that other processes are acting to clear the mantle or
to keep it from completely choking off the activity in the rst
place, but this appears to be less common than the mantling
process itself. In any case, it has become clear that understand-
ing the nature of the active areas is an important step in under-
standing the evolutionary processes that act on cometary nuclei.
Studies of the coma morphology and how it changes as a
function of time provide information toward understanding the
active areas. Dust features are particularly revealing because
they tend to have a higher signal-to-noise than the gas features,
and, because the dust maintains a relatively predictable trajec-
tory after decoupling from the gas, the dust features represent a
long-term record of emission that can be followed to large dis-
tances. The gas species provide an additional constraint on the
activity but involve additional complications, including molec-
ular dissociation that changes the composition of the outowing
material and leads to random excess velocities. When used in
conjunction, however, studies of the gas and dust morphology
can provide strong constraints on the nucleus activity. After the
source regions have been identied, they can be studied to de-
termine their physical properties and to investigate how they
respond to the solar insolation that drives the comets activity.
Identication of the sources and characterization of their prop-
erties also provides constraints that can be used in other studies
relating to seasonal effects, nongravitational forces, chemical
inhomogeneities, or any other phenomenon relating to jet activ-
ity and the morphology of the dust or gas coma (
Feaga et al.,
2007; McFadden et al., 2006
).
As the Deep Impact (DI) spacecraft approached Comet Tem-
pel 1 for its 4 July 2005 yby and impact experiment, it ob-
tained thousands of observations of the coma and nucleus. The
temporal coverage was unprecedented, with images obtained
every 4 h for nearly 2 months before encounter. This sequence
is invaluable for revealing the temporal changes in the coma
and illustrating how details become clearer and more obvious
as the spatial resolution improves. Early images showed a sim-
ple asymmetric coma with little structure, but as the spacecraft
closed on the comet, more and more features became apparent
and were observed to change with time. Not only was the tem-
poral coverage of the comet unprecedented, but the inner coma
and nucleus were observed at a spatial resolution higher than
ever before achieved. In the highest-resolution images, intri-
cate structures are visible in the inner coma, at a level of detail
never seen before. The asymmetries and features in the coma
are most likely produced by active areas on the nucleus, which
may consist of small, isolated spots or larger extended regions.
The wealth of observations from the DI encounter presents us
with an excellent means of exploring the connection between
the coma and the active sources on the nucleus, to determine
where the jets originate, to study the characteristics of the dif-
ferent sources, and to investigate the more general issue of
mantling processes on the surface.
In this work, we present an overview of the DI dust coma ob-
servations that were obtained during the mission. We describe
the morphology and how it changed throughout the approach
and encounter with the nucleus and highlight some of the in-
teresting features that were seen. We also present results from
some preliminary measurements and outline the future analy-
ses that will be done to bring the different aspects of the coma
morphology into a broader general context.
2. Observations and image enhancements
The results presented here were derived from data obtained
with the Medium Resolution Instrument (MRI) and the High
Resolution Instrument (HRI) cameras aboard the DI yby
spacecraft and the Impactor Targeting System (ITS) camera on
the impactor (
Hampton et al., 2005
). These images have a high
dynamic range (
5 � 10
3
)
that allows bright nucleus features
to be seen in the same frame in which coma features are visi-
ble off the limb. Occasional long exposures were also taken to
improve the S/N of the coma. These long exposures tended to
saturate the nucleus and inner coma, but they were typically ob-
tained in conjunction with a shorter exposure that can be used
to ll in for the saturated regions. For studies of the dust coma,
we use the unltered images from the ITS and the MRI and HRI
images obtained using the broadband context lters (
700 nm
wide, centered at
650 nm). Because these lters are broad,
the observations are dominated by the sunlight reected off the
dust grains, producing high signal-to-noise images of the dust
while minimizing the contamination from gas emissions. Ba-
sic processing of the images was done using the standard DI
science pipeline, described by
Klaasen et al. (2007)
. It included
bias removal using the CCD overscan regions, dark subtraction,
at elding, corrections for frame transfer smear and conver-
sion to absolute uxes. Where appropriate, information about
the specic observations is included with the individual discus-
sions below.
Fig. 1
shows an example of an MRI coma/nucleus image
(left) and a high-resolution ITS image showing the detailed fea-
tures on the nucleus (right). The coma image is formed from
two separate frames: 9000852 is a 2-s exposure that captured
the coma with a high signal while saturating the brighter nu-
cle