Fig' 3' Model residual map for DS1 image at 51'6 relative to the raw image' The RMS residual is abou

1
Surface Photometric Variation of Comet Borrelly's
Nucleus
Jianyang Li, Michael F. AHearn, Lucy A.
McFaddenDepartment of Astronomy, University of
Maryland, College Park, 20742 email
jyli_at_astro.umd.edu
Introduction Comet Borrelly was visited
by Deep Space 1 (DS1) in Sept., 2001 (Soderblom
et al. 2004) The images of comet Borrelly's
nucleus show large brightness variation over the
surface even after the effect of shape is taken
into account (Oberst et al. 2004, Kirk et al.
2004). Oberst et al. (2004) and Buratti et al.
(2004) interpreted it as albedo variation, but
Kirk et al. (2004) concluded that roughness
variation causes reflectance variation. Beginning
with the terrain partition from Britt et al.
(2004), we created partitions based on brightness
and phase ratio of brightness. We determined
Hapke's (1993) photometric parameters for each
terrain unit separately and conclude that the
brightness variation among the geological
terrains is caused by a combination of variations
in all of single scattering albedo, surface
roughness, and single-particle scattering
asymmetry. The data we used include the
last 10 DS1 MICAS image with Borrellys disk
resolved, covering phase angle from 51.6? to
75.7?. Resolution changes from 46.7 m/pix to
122.7 m/pix. To calculate the incidence and
emission angles, we used Borrellys shape model
from Kirk et al. (2004), as well as the SPICE
data for DS1 spacecraft.
Fig. 5. Maps for three photometric parameters of
comet Borrelly. From left to right are, single
scattering albedo (SSA), asymmetry factor (g) of
single-particle phase function, and roughness
parameter (?), respectively. Below each map is
the corresponding histogram. Their averages are
marked by dotted lines.
Fig. 4. Modeled phase ratio of 59.6?/51.6?. The
color table for this figure is similar to that
used in Kirk et al. (2004) phase ratio plot
(left). As same color represents same value as
in Kirks figure, they can be compared directly.
Fig 1. Our photometric terrain definition
overlapped on DS1 image (left panel) and
Borrellys reflectance ratio of 59.6?/51.6?
(right panel, from Kirk et al. (2004)).
Photometric terrains are modified from Britt et
al. (2004), and defined for both different
brightness and different radiance phase ratios.
Fig. 3. Model residual map for DS1 image at
?51.6? relative to the raw image. The RMS
residual is about 18. Large error mostly appears
close to the edges of terrains, probably due to
the misalignment of terrain model with image,
and/or the large photometric variation at terrain
edges.
Fig 6. Scatter plot of photometric parameters of
terrains on comet Borrelly. Horizontal axis is
the SSA, vertical axis is the asymmetry factor
(g). Colors of symbols indicate the roughness of
terrain as in the color-bar at right-hand side.
Sizes of symbols indicate the projected area
fractions of terrains in the DS1 image with
?51.6?.
Fig. 7. The polar day region on the surface of
Borrelly at the time of DS1 flyby is marked by
white color. The upper left end is in polar day
region, but the upper right end is not.
Therefore, the upper right end must have activity
periodically turned on and off, as observed 0.4T
earlier by DS1. Thus high roughness is correlated
with fan jet activity, as at the lower end.
Fig. 2. An examples of the Hapkes photometric
fitting. This fit is for the upper right end
terrain, with 12 RMS residual. Left panel shows
the fitted value plotted as a function of
measured value. Perfect correlation will result
in the solid line in the plot. Right panel is
the ratio of modeled value to measured value as a
function of incident angle, emission angle, and
phase angle, respectively. Fittings to other
terrains are typically worse than this one, with
RMS ranging from 10 to 23.
Conclusions The reflectance variation of
comet Borrelly is likely due to the combination
of single-scattering albedo (SSA) variation, the
variation of asymmetry factor (g), and roughness
(?) variation. The SSA ranges from 0.03 to 0.08,
the g is from nearly isotropic scattering (-0.1)
to strongly back-scattering (-0.7), and some
terrains are very smooth (?lt10?) but some are
very rough (?gt50?). The photometric
difference of terrains may have evolutionary
origin or be caused by cometary activities. From
diurnal variation of local solar elevation angle
on Borrellys surface, we correlate high
roughness region with fan jet activities, which
also possibly correlates with their low
back-scattering particle phase functions, and
high albedos. However, more and better
photometric data are needed to confirm this
conclusion, to understand the physical processes
caused by jet activities on the surface, and to
track the photometric evolution of cometary
surface.