[meteorite-list] Deep Impact Mission Update - February 2006

From: Ron Baalke <baalke_at_meteoritecentral.com>
Date: Tue Feb 21 11:50:15 2006
Message-ID: <200602211608.k1LG8vh18759_at_zagami.jpl.nasa.gov>

http://deepimpact.jpl.nasa.gov/mission/update-200602.html

Deep Impact Mission Update
January/February 2006

Water Ice Found on a Small Portion of the Comet's Nucleus
By Lucy McFadden

In a paper appearing in Science Express on Feb. 2, 2006, an article by
Sunshine et al. reports on the Deep Impact science team's finding of a
small area of water ice on the surface of Tempel 1. This is the first
time that water ice has been observed on the surface of a comet. Past
efforts with the near-IR spectrometer on Deep Space 1 mission flying
past comet Borrelly and from the ground of comets far from the sun and
not enshrouded with coma, have yielded no evidence of water ice on their
surface.

As a comet approaches the Sun, it releases gas and dust in its immediate
vicinity forming the coma and obscuring the nucleus from view unless
spacecraft can get at close range. Deep Impact did just that. Imaging
with the two cameras, the HRI and MRI showed small regions that were
about 30% brighter than surrounding areas. After scaling the images to
an average value of the nucleus, three discrete areas on the nucleus are
brighter in the ultraviolet and darker in the near-infrared. When
Co-Investigator Dr. Jessica Sunshine looked at the spectra in that
region, after subtracting a thermal component, what was left was the
spectral signature of water ice, in the form of absorption bands at 1.5
and 2.0 ?m. Absorption bands at these wavelengths are diagnostic of
water ice. The combination of the relative colors and the spectra make a
powerful case that there is water ice at these specific locations on
Tempel 1.

Given that the spectrometer has a two dimensional detector, it is
possible to make a map of Tempel 1 at the wavelength of the ice
absorption bands. That map shows that the bright regions in the UV are
correlated with dark regions in the near-IR where water ice absorbs
light. Since the visible images have a higher spatial resolution, we use
those images to calculate the extent of ice on Tempel 1's surface. That
turns out to be a small fraction of the surface, only 0.5%. Next, the
temperature map is combined with the color map, showing that two of the
three regions are colder regions of the nucleus. Stereo images show the
largest area of ice to be a depression 80 meters below surrounding
areas. Never the less, the temperatures in this region are 285 -295 K,
significantly above the ~200K at which ice would sublimate in space at
the location of Tempel 1.

What is significant is that the extent of this ice on Tempel 1's surface
is not sufficient to produce the observed abundance of water and its
by-products in the comet's coma. The team thus concludes that there are
sources of water from beneath the comet's surface that supply the
cometary coma as well.

Also important is that the particle size of the water ice, is greater
than the icy grains in the coma, and is probably recondensed onto the
comet's surface. It is therefore probably not a primary block of
cometary material which would be called a cometesimal.

------------------------------------------------------------------------

Science Results, the View from Ground and Space
By Ray Brown

Introduction
Is it all over but the shouting? Definitely, not. It is time to use the
massive amount of data relayed back to Earth from the Deep Impact
instruments. Analysis will likely go on for years.

This article is a digest of a set of papers prepared by the Deep Impact
science team and collaborators and published in a special section of the
October 14, 2005 issue of Science. Here we focus on background science,
results and conclusions rather than data or analytical methods. In
particular we study the results relating to our search for primordial ices.

K. J. Meech et al. present a summary of observations made world-wide and
in space. They point out that observations began in 1997 and continued
into 2005, and that, in 2005 alone, the campaign claimed the attention
of 73 Earth-based telescopes at 35 observatories in addition to
observatories based in space. It was "an unprecedented coordinated
observational campaign." [1]

Figure 1
Fig. 1: Map of Earth, showing the locations of observatories
collaborating in the coordinated campaign (red dots). Excerpted with
permission from K. J. Meech et al., Science 310, 265 (2005); published
online 8 September 2005 (10.1126/science.1118978). Copyright 2005 AAAS
<http://www.sciencemag.org>.

Before, During and After with the Keck 2
The telescopes in Hawaii had the second best seat (after the flyby
spacecraft) for watching the encounter. In their article, M. J. Mumma et
al.[2] describe their observations made with the 10 meter
Keck 2 telescope high atop Mauna Kea
in Hawaii.

As the stream of ejecta emerged from the crater its intensity increased
due to the outflow of dust. and also its ability to emit light. The
intensity of the total coma rose quickly for about 40 minutes after
which it rose slowly and leveled off toward the end of observing time,
about 2 hours after impact, see top curve in Fig. 2. Peak intensity of
the coma was measured close to the nucleus. It also rose quickly but
only for 15 minutes after which it declined to its pre-impact level
about 90 minutes after impact, see red curve in Fig. 2.

Figure 2
Fig. 2: Light curves obtained from the SCAM images (black) and from the
spectral continuum (3.3 ?m) in individual spectra. All light curves show
a rapid rise of intensity after impact. After its maximum, the peak
spectral intensity (red) falls rapidly to its preimpact value by UT
7:20. The total spectral intensity (blue) decays more slowly, and the
total coma intensity (black) plateaus.
Excerpted with permission from M. J. Mumma et al., Science 310, 270
(2005); published online 15 September 2005 (10.1126/science.1119337).
Copyright 2005 AAAS <http://www.sciencemag.org>.

When scientists analyze spectroscopic data they often look at two
quantities derived from their data. The column number is the total
number of molecules in a column viewed by the spectrometer. The relative
abundance of a molecule with respect to water is the column number of a
molecule of interest divided by the column number of water. Relative
abundances are important because they provide a way to compare
conditions at different times, before impact with after impact for example.

The high dispersion spectrometer on the Keck 2 telescope captured the
spectra of eight gases in the ejecta: water, ethane, hydrogen cyanide,
carbon monoxide, methanol, formaldehyde, acetylene and methane. The
chemical symbols for these are respectively H2O, C2H6, HCN, CO, CH3OH,
H2CO, C2H2, and CH4.

Only water, ethane, methanol and hydrogen cyanide were measured both
before and after impact. After impact the abundance of methanol and
hydrogen cyanide remained unchanged. However, the abundance of ethane
was enhanced either by a factor of 1.8 or 3.0 depending on the method of
analysis. The different values may be due to pre-impact outgassing being
computed in different ways.

Three molecules have low sublimation temperatures and are therefore
likely to be less prevalent in the nucleus. They are carbon monoxide,
methane and ethane. It is suggested that because two of these, carbon
monoxide and methane, were not measured in the coma before impact, they
have been sublimated out of the near surface region by the heat of the
sun. Keck, on the other hand, measured ethane before impact, and in
greater abundance, after impact.

The View from Space, Rosetta, OSIRIS

On July 4th, ground based telescopes, large and small, all over the
world were trained on Tempel 1. Meanwhile, the OSIRIS cameras on board
the Rosetta spacecraft also recorded the impactor's collision. Rosetta
is a European Space Agency craft en route to comet 67P/Churyumov-Gerasimenko.
Results are reported in an article by H. U. Keller et al.[3]
Although the OSIRIS cameras also observed other gas components of the
coma and ejecta, the report in Science focuses largely on the radical CN.

Figure 3
Fig. 3: The Comet 9P/Tempel 1 as seen by the OSIRIS Narrow Angle Camera
System on board ESA's Rosetta comet-chaser spacecraft mission, on 30
June 2005, three days before the Deep Impact encounter. The distance
between Rosetta and Tempel 1 was 80 million kilometres. The stars seen
in the image are elongated because the Rosetta spacecraft was actively
tracking on the moving comet while the image was acquired. Credit:
ESA/OSIRIS consortium

A main objective of the Deep Impact mission is to discover differences
between the near surface composition of Tempel 1 and the composition of
its interior. To accomplish this, scientists on the Rosetta team
analyzed the production of the radical CN. CN is a negatively charged
ion based on a carbon atom and a nitrogen atom. It is produced when a
more complex molecule, its parent molecule, decomposes. HCN, hydrogen
cyanide, is probably a parent molecule of CN. Fig. 4 shows how the
number of CN molecules rises with time.

CN molecules vs. time, from OSIRIS data.

Figure 4
Fig. 4: Number of impact-created CN molecules as a function of time. The
data for apertures of different radii (1 pixel = 31,200 km) are compared
with models showing approximate lower (4 x 1029) and upper limits (6 x
10 29) for the number of parent molecules created at the time of impact.
Excerpted with permission from H. U. Keller et al., Science 310, 281
(2005); published online 8 September 2005 (10.1126/science.1119020).
Copyright 2005 AAAS <http://www.sciencemag.org>.

Through a series of steps that estimate the rate at which CN and HCN
molecules are produced, the Rosetta team was able to estimate the total
number of molecules of both CN-parents and CN itself. The number of
water molecules was also determined and the ratio of CN parent-molecules
to water molecules computed for conditions before, during and after the
impact event. The ratios suggest there was a greater abundance of CN
parent molecules in the impact produced ejecta cloud than in Tempel 1's
coma prior to impact. That can be interpreted to mean that while the
comet's normal outgassing originates near the surface, Deep Impact has
excavated HCN-enriched material from farther below.

Origins and Classifying Comets

A clear, concise description of comet formation and distribution in the
solar system appears in the article by Mumma et al. According to current
thinking, there is a disk of comets and other objects lying in the plane
of the solar system that flares out until it becomes a spherical shell
enclosing the entire solar system. The disk, called the Kuiper belt,
begins at about the orbit of Neptune and extends out for several hundred
astronomical units, where an astronomical unit is about 93 million
miles. Pluto is considered by some to be the innermost Kuiper belt
object. The shell, called the Oort cloud, is humongous. It occupies,
roughly, the region from 10 thousand to 50 thousand astronomical units
from the sun.

We now turn our attention to how and where comets formed. It may be that
gravity caused unevenly distributed gas and dust grains in a giant
molecular cloud to draw together. Gases, such as water vapor, and carbon
dioxide froze to become the pristine ice that we seek so much. The
freezing ices coated colliding dust grains and cause them to stick
together forming clumps of ice and dust. Then the frozen clumps,
collided to form small comets called cometesimals, which in turn
collided and form the large comets that we see today. We would very much
like to know in what region or regions of the solar system comets
formed. Current thinking is that some comets, such as Tempel 1, formed
in the region beyond where Jupiter orbits today, were perturbed into the
Kuiper belt and the Oort cloud by the gravity of the major planets and
then perturbed again back into the inner solar system.

After comparing abundances of molecules in Tempel 1, Mumma et al.
suggest that, to paraphrase slightly, Tempel 1 and most comets in the
Oort cloud formed in the same region of the protoplanetary disk. They
add that their suggestion is consistent with the idea that comets that
have been perturbed into the Kuiper belt and Oort cloud comets
originated in the outer giant planets' region of the protoplanetary disk.

We can classify comets by the reservoir from which they entered the
inner solar system, Oort cloud or Kuiper belt. We can also classify them
by the characteristics of their orbits. For example, Jupiter family
comets can be classified by how close they come to the sun and how they
behave when encountering Jupiter. When classified in this way, Tempel 1
is a Jupiter-family comet. However, Meech et al. point out that some
molecules excavated by the impact have abundances consistent with
abundances of typical Oort cloud comets. Those molecules are water,
ethane, methanol, acetylene and hydrogen cyanide.

References

1. K. J. Meech et al., Science 310, 265 (2005); published online 8
September 2005 (10.1126/science.1118978).

2. M. J. Mumma et al., Science 310, 270 (2005); published online 15
September 2005 (10.1126/science.1119337)

3. H. U. Keller et al., Science 310, 281 (2005); published online 8
September 2005 (10.1126/science.1119020)

------------------------------------------------------------------------

Deep Impact Wake-Up Status
By Mike A'Hearn

Dear DI Science Team,

This is just to let you know that Jenn Rocca called me yesterday evening
[Feb 9] to say that the initial reports from the DI wake-up activity
were looking good. She then sent out during the night her Flight
Director's Report. Everything they have tested has been fully functional
and in the states that were expected. Thus the spacecraft seems healthy
for an extended mission.

This morning they will bring the spacecraft to point state, i.e. bring
it out of the safe mode that it has been in since August.
Received on Tue 21 Feb 2006 11:08:56 AM PST


Help support this free mailing list:



StumbleUpon
del.icio.us
reddit
Yahoo MyWeb