MSI C/1996 B2 (Hyakutake) observation plans
NEAR MULTISPECTRAL IMAGER C/1996 B2 (Hyakutake) OBSERVATION PLANS
Jim Bell, Joe Veverka, Ann Harch (Cornell University)


 href=http://newproducts.jpl.nasa.gov/comet/hyakutake/index.html
img align=bottom SRC=gif/near/hyakutake/hanon6.sm.gif
Link to JPL's C/1996 B2 Home Page 


Updated on 16 March 1996 at 12:30 am EST</blink>

MSI/NIS Team Members,

This is the latest updated version of the MSI Hyakutake
image sequence.  

This update includes new, accurate estimates of the RA, DEC, and Sun
angle of each of our mosaic positions.  These values were provided
by David Dunham.  The cometary orbit was found from Don Yeomans'
Comet 1996 B2 (Hyakutake) orbital solution #14, using 342 observations 
made from Jan. 1 through March 13; the geocentric angular uncertainty is 
expected to be only 0.008 deg. during the close approach on March 24-25.  
The NEAR orbit used was solution ntcm2n0a, JPL's latest post-TCM1
spacecraft orbit.

This version of the plan incorporates the total 15 Mbit data volume
limit as well as an increased appreciation of the fact that
we are only "guaranteed" a downlink of the first 4.5 Mbits;
anything more than that will only be sent if downlink resources
allow and if the additional downlink does not interfere with 
other planned MSI calibration operations in April.

Highlights: We get 28 of the 45 images in the "original"
sequence. What we have given up is the third image at
each color/position and the darks at the pre-comet temperatures.
We have also incorporated a higher degree of risk by relying
more on compression.  

The first six images nearly fill the 4.5 Mbits allowed for
"immediate downlink" (this will take nearly 30 hours), and
are judged "highest priority".  The sequence has been slightly 
modified by moving one each of the filter 1 and filter 2 images
into this highest priority slot so that some color information
will be returned in the initial priority downlink.  To fit these
images in the top 5, one of the darks has been moved to the
end of the sequence.  If possible, 
the other 22 images will come trickling down over the several week 
period after comet encounter but before the major April calibration
sequence begins.  The other reason
for moving a few images around was the desire expressed by Ed Hawkins
and Rob Gold to keep similarly-compressed images grouped together.  
Ed and Rob feel that additional filter wheel moves are not an
important consideration compared to the extra burden of continually
changing the compression commands.  

Compression is estimated to yield a factor of 2.5 compression
per image using the lossless Fast scheme, and a factor of 
2.5*1.5=3.75 compression using the linear-log 12-to-8-bit 
compression (Table 2) followed by Fast compression.  These numbers
may be a bit conservative depending on what the comet actually
looks like, and so the sequence contains a little padding
beyond 15 Mbits in case the images compress further than we
predict.  If the images do not compress as much as we predict,
then the download will simply be cut short somewhere in
one of the farther tail positions, which are lowest priority.  

This pared-down scheme will still allow us to get inner coma
morphology, color (including near-IR), a large coma mosaic,
and a hefty chunk of the tail.  The latest data I've seen
still indicate that the comet will reach integrated magnitude
of +1 to 0.  For scale, if this flux were spread out evenly
over a 10,000 km region (which it is not, but this is a
fair approximation to a ~1 million km coma with a r**-2
brightness gradient), or a 6x3 MSI pixel region at the
time of encounter, then in 999 ms in filter 0 with the 
cover on, MSI would detect from 533 to 1332 DN/pixel.
This corresponds to 19 to 47 DN/pixel in filter 2,
59 to 147 DN/pixel in filter 1, and 55 to 139 DN/pixel
in filter 5.  These are most likely upper limits,
but it's impossible to know whether the brightness will
surge or fade in the next several weeks.  I have not 
seen any reliable estimates of the tail magnitude, so
these tail images may be a total bust unless we get lucky.  

Many thanks to Brian Carcich, Jonathan Joseph, Bill Owen, Jim Miller, 
Don Yeomans, Lucy McFadden, Mike Malin, David Dunham, and Scott Murchie 
for substantial help on this!


<b>COMET HYAKUTAKE MSI OBSERVATION SUMMARY:</b>

Remember the ground rules: Don't
let the solar panels get more than 10 degrees from the Sun, and
keep the data volume to approximately 15 Mbits.  We have
chosen March 24 around 15:00 UT for our sequence as that is the
time when the solar panel/Sun angle is near zero.  This time also falls within 
the scheduled DSN pass for that day, which is from 08:15 to 16:30 UT. We have
tried to fold in as much information on the comet as we can based on
the latest WWW reports and IAU circulars.  However, comets are
precocious beasts, and so things (size, brightness, extent of tail) 
could change over the next few weeks.  It is not clear whether we will
have the opportunity to tweak the exposure times once more using
more updated comet information as we get closer to the time of observations.

OUTLINE:

Image the coma with a 2x2 mosaic plus an extra position centered exactly on 
the coma, and image the tail with a 1x4 mosaic 
(<A href=gif/near/hyakutake/fig1.gif>Figure 1</). 

<img align=bottom SRC=gif/near/hyakutake/fig1.gif> FIGURE 1: Schematic representation
of MSI C/1996 B2 image sequence.  The sequence begins with images centered
on the coma, then proceeeds to a 2x2 mosaic centered on the coma, and ends
with a 1x4 mosaic down the tail.  The scalebar in km is correct but the 
overlap and phase information in the caption next to the scale is inaccurate. 
The spacecraft twist is such that at this time North is approximately straight up in 
these images.  The coma is modeled as a sphere of diameter = 1,000,000 km and the 
tail is modeled as a cylinder of length = 15,000,000 km.  The tail is oriented in the 
anti-sun direction. Each MSI field of view is approximately 973,000 x 743,000 km.



SEQUENCE OUTLINE: 



1) Startup and Initial Dark.</  Start up MSI at day 84/08:45 UT (March 24, 1996).
At around 84/14:45 UT, slew to a field near the comet with no bright stars. 
We recommend [RA, DEC] = [108.925, 61.80]. (Sun angle off the z-axis is 2.16 degrees).
Slew and settling must be completed by 84/14:50 UT.  Go to filter 7. 
At 84/14:50 UT acquire one 999 ms dark frame for calibration of the comet images.


   Frame 01  Exp=999 ms  Filt=7  Compression=Linear-Log + Fast (long dark at T = -30)




2) Coma Nadir Clear: No Compression. At 84/14:55 UT, slew camera boresight to comet 
nadir position.  David Dunham's calculations indicate that the boresight position (at 14:55 UT) 
will be RA=105.31, DEC=67.84 (J2000) as seen from the spacecraft, and the Sun angle will be 
0.33 degrees off the z-axis.  Slew and settling must be complete by 84/15:00 UT. Go to filter 0.  
At 84/15:00 UT acquire a single, uncompressed clear filter coma image:


    Frame 02 Exp=999 ms  Filt=0  Compression=None  (Coma-centered clear images)




3) Coma Nadir Clear and Color: Lossless Compression. Immediately after sequence step 2, take 
the following images, all using FAST compression: 


    Frame 03 Exp=250 ms  Filt=0  Compression=Fast  (Insurance against coma saturation)
    Frame 04 Exp=999 ms  Filt=1  Compression=Fast  (550 nm coma images)
    Frame 05 Exp=999 ms  Filt=2  Compression=Fast (450 nm coma images)
    Frame 06 Exp=999 ms  Filt=0  Compression=Fast
    Frame 07 Exp=500 ms  Filt=0  Compression=Fast
    Frame 08 Exp=999 ms  Filt=5  Compression=Fast (900 nm coma images)




4) Coma Nadir Color: Lossy Compression. Immediately after sequence step 3, take the following 
images, all using FAST & LOSSY Table 2 (LINLOG) compression: 


    Frame 09 Exp=999 ms  Filt=1  Compression=Linear-Log + Fast
    Frame 10 Exp=999 ms  Filt=2  Compression=Linear-Log + Fast
    Frame 11 Exp=999 ms  Filt=5  Compression=Linear-Log + Fast 




5) Coma 2x2 mosaic: Lossless Compression.  As soon as the previous sequence step is finished, slew camera 
boresight to elevation 2.22 degrees, azimuth 203.9 degrees relative to nadir (Elevation and azimuth are 
measured in degrees from nadir. Zero azimuth is Sun-pointed, and azimuth increases clockwise).  
Go to filter 0.  Slew and settling must be complete by 84/15:10 UT.
Predicted boresight position is RA = 110.70, DEC = 68.34, Sun angle = 1.56 deg. 
Acquire the following images for the NW quadrant of the coma mosaic: 


    Frame 12 Exp=999 ms  Filt=0  Compression=Fast (Coma NW mosaic position)
    Frame 13 Exp=999 ms  Filt=0  Compression=Fast
 

Then slew camera boresight to elevation 1.07 degrees, azimuth 302.8 degrees.
Slew and settling must be complete by 84/15:15 UT.
Predicted boresight position is RA = 103.52, DEC = 68.74, Sun = 1.09 deg.
Acquire the following images for the NE quadrant of the coma mosaic: 


    Frame 14 Exp=999 ms  Filt=0  Compression=Fast  (Coma NE mosaic position)
    Frame 15 Exp=999 ms  Filt=0  Compression=Fast
 
Then slew camera boresight to elevation 1.07 degrees, azimuth 57.2 degrees.
Slew and settling must be complete by 84/15:20 UT.
Predicted boresight position is RA = 102.94, DEC = 66.94, Sun = 1.14 deg.
Acquire the following images for the SE quadrant of the coma mosaic: 


    Frame 16 Exp=999 ms  Filt=0  Compression=Fast  (Coma SE mosaic position)
    Frame 17 Exp=999 ms  Filt=0  Compression=Fast
 
Then slew camera boresight to elevation 2.22 degrees, azimuth 156.1 degrees.
Slew and settling must be complete by 84/15:25 UT.
Predicted boresight position is RA = 109.36, DEC = 66.55, Sun = 1.42 deg.
Acquire the following images for the SW quadrant of the coma mosaic: 


    Frame 18 Exp=999 ms  Filt=0  Compression=Fast  (Coma SW mosaic position)
    Frame 19 Exp=999 ms  Filt=0  Compression=Fast




6) Tail 1x4 Mosaic: Lossy Compression.  Slew camera boresight to elevation 
2.5 degrees, azimuth 180 degrees relative to nadir.
Slew and settling must be completed by 84/15:30 UT. 
Predicted boresight position is RA = 110.89, DEC = 67.32, Sun = 1.84 deg.
Acquire the following images for the first tail mosaic position: 


   Frame 20  Exp=999 ms  Filt=0  Compression=Linear-Log + Fast  (Tail position 1)
   Frame 21  Exp=999 ms  Filt=0  Compression=Linear-Log + Fast

Then slew camera boresight to elevation 5.0 degrees, azimuth 180 degrees.
Slew and settling must be completed by 84/15:35 UT. 
Predicted boresight position is RA = 116.93, DEC = 66.65, Sun = 4.30 deg.
Acquire the following images for the second tail mosaic position: 


   Frame 22  Exp=999 ms  Filt=0  Compression=Linear-Log + Fast  (Tail position 2)
   Frame 23  Exp=999 ms  Filt=0  Compression=Linear-Log + Fast

Then slew camera boresight to elevation 7.5 degrees, azimuth 180 degrees.
Slew and settling must be complete by 84/15:40 UT.
Predicted boresight position is RA = 122.60, DEC = 65.76, Sun = 6.75 deg.
Acquire the following images for the third tail mosaic position: 


   Frame 24 Exp=999 ms  Filt=0  Compression=Linear-Log + Fast  (Tail position 3)
   Frame 25 Exp=999 ms  Filt=0  Compression=Linear-Log + Fast

Then slew camera boresight to elevation 10 degrees, azimuth 180 degrees.
Slew and settling must be complete by 84/15:45 UT.
Predicted boresight position is RA = 127.86, DEC = 64.67, Sun = 9.20 deg.
Acquire the following images for the fourth tail mosaic position: 


   Frame 26 Exp=999 ms  Filt=0  Compression=Linear-Log + Fast  (Tail position 4)
   Frame 27 Exp=999 ms  Filt=0  Compression=Linear-Log + Fast
 



7) Second Dark.  At around 84/15:50 UT, slew back to the field near the comet 
with no bright stars. ([RA, DEC] = [108.925, 61.80]). Go to filter 7. 
Acquire one 10 ms dark frame for calibration of the comet images.


   Frame 28  Exp= 10 ms  Filt=7  Compression=Linear-Log + Fast (short dark at T = -30)




0) Sequence 0.</  This is the "don't change anything" sequence that is
supposed to get tacked onto the end of everything.  I don't think that we will
download this image; its main purpose, I believe, is to tidy up the instrument 
and leave it in a state where it can sit happily idle for a while.  


   Frame 29  Exp= 10 ms  Filt=7  Compression=Linear-Log + Fast (short dark at T = -30)


                 
* Notes:

1. Frames taken at the same position should be acquired as rapidly as possible 
(minimum separation time of 3 to 5 sec acceptable). 


2. We weren't sure what settling time to assume.  We estimate that each mosaic 
positioning slew (~2.5 degrees) will take about a minute.  If that's correct, that 
leaves about 4 minutes for settling time for each position in this sequence. 


3. Our model uses the most current SPICE data available for the spacecraft 
trajectory.  The SPICE data for the comet were obtained from an approximate comet 
ephemeris generated by Don Yeomans.  The comet's coma is modeled as a sphere of 
diameter = 1,000,000 km and the tail is modeled as a cylinder of diameter = 10,000 km, 
and length = 15,000,000 km.  The tail is oriented in the anti-sun direction.  The 
coma is assumed to have a total integrated Vmag = +1. 


4. Attached is a star plot for the coma-centered mosaic position. Two stars will be 
in the coma-centered field-of-view at 84/15:00 UT.  They are BS2511 (PPM15946) and 
BS2490 (PPM15925) according to information provided by Bill Owen (JPL Navigation Team). 
Two other stars (BS2830=PPM16214 and BS2809=PPM16193) will be in the field of view of
the 2x2 coma mosaic as well.
(<A href=gif/near/hyakutake/fig2.gif>Figure 2</). 

<img align=bottom SRC=gif/near/hyakutake/fig2.gif> FIGURE 2: Map of bright stars in
the MSI field of view (inner rectangle) for March 24 at 15:00 UT.  The numbers and
letters for each star correspond to the PPM catalog number, the V magnitude times 10, 
and the spectral type.  Figure provided by Bill Owen of the JPL Navigation Team.



5. Unless otherwise noted, define sequences and execute at appropriate MET's.


6. Send down summary images as well as full images, to check pointing and consistency
of header information.


7. When moving the filter wheel, wait > 1 sec after end of last exposure.


8. 28 frames total not counting last sequence 0 image; Approximately 15 Mbit data 
volume (exact volume depends on actual compression factors).  

The following table shows the cumulative number of bits predicted for this sequence,
using (overly conservative??) compression estimate numbers provided by Rob Gold: 



MSI C/1996 B2 Image sequence
Summary and cumulative bit totals

#  Mbits  Exp  Filt  Compression     Description 
-- -----  ---  ----  -----------     ---------------------------------------
01  0.46  999   7    Log+Fast 3.75x  long dark at T = -30
02  2.14  999   0    None 1x         Coma-centered clear image
03  2.82  250   0    Fast 2.5x       Insurance against coma saturation
04  3.49  999   1    Fast 2.5x       550 nm coma image
05  4.17  999   2    Fast 2.5x       450 nm coma image
06  4.84  999   0    Fast 2.5x       Coma-centered clear image

07  5.52  500   0    Fast 2.5x       Insurance against coma saturation
08  6.20  999   5    Fast 2.5x       900 nm coma image
09  6.65  999   1    Log+Fast 3.75x  550 nm coma image
10  7.11  999   2    Log+Fast 3.75x  450 nm coma image
11  7.57  999   5    Log+Fast 3.75x  900 nm coma image
12  8.24  999   0    Fast 2.5x       Coma NW mosaic position
13  8.92  999   0    Fast 2.5x       Coma NW mosaic position
14  9.60  999   0    Fast 2.5x       Coma NE mosaic position
15 10.27  999   0    Fast 2.5x       Coma NE mosaic position
16 10.95  999   0    Fast 2.5x       Coma SE mosaic position
17 11.62  999   0    Fast 2.5x       Coma SE mosaic position
18 12.30  999   0    Fast 2.5x       Coma SW mosaic position
19 12.98  999   0    Fast 2.5x       Coma SW mosaic position
20 13.43  999   0    Log+Fast 3.75x  Tail position 1; 2.5 deg. from coma
21 13.89  999   0    Log+Fast 3.75x  Tail position 1; 2.5 deg. from coma
22 14.35  999   0    Log+Fast 3.75x  Tail position 2; 5 deg. from coma
23 14.89  999   0    Log+Fast 3.75x  Tail position 2; 5 deg. from coma

24 15.26  999   0    Log+Fast 3.75x  Tail position 3; 7.5 deg. from coma
25 15.72  999   0    Log+Fast 3.75x  Tail position 3; 7.5 deg. from coma
26 16.17  999   0    Log+Fast 3.75x  Tail position 4; 10 deg. from coma
27 16.63  999   0    Log+Fast 3.75x  Tail position 4; 10 deg. from coma
28 17.09   10   7    Log+Fast 3.75x  short dark at T = -30
29 17.55   10   7    Log+Fast 3.75x  short dark at T = -30
</pre>




Last Modified by <A HREF = "Welcome.html">Jim Bell</a> on 13 March 1996.
Mail to: <A HREF="mailto:jimbo@cuspif.tn.cornell.edu"><jimbo@cuspif.tn.cornell.edu</</