molczan@gpu.utcs.utoronto.ca (Ted Molczan) (11/13/90)
STS 38 Visual Observation Guide
-------------------------------
by T.J. Molczan, Toronto, Canada
12 Nov 1990
This report is intended to assist those who wish to make visual observations
of STS 38. This is a DoD mission, and therefore, most aspects of the mission
have been classified, including the orbit. However, it is possible to make
an accurate estimate of the orbital elements using the information that is in
the public domain. Use has been made of basic orbital mechanics and some
leaked information made available by various news media.
Highlights:
-----------
Section 2 provides a simple step by step procedure to produce an estimate of
the orbital elements in the standard NORAD "2-line" format. Section 3 has
tables that tell you whether or not the shuttle will be visible at your
latitude. Section 5 includes information on obtaining free software for
making orbital predictions and how to make and share observations.
1.0 STS 38 Mission Synopsis
-----------------------
1.1 AV WEEK Report
--------------
According to AVIATION WEEK and SPACE TECHNOLOGY, STS 38 will launch a digital
imaging reconnaissance satellite, with a mass of about 10,000 kg.
The shuttle will enter a 217 km orbit at an inclination of 28.45 deg to the
equator. It will later raise its altitude to at least 241 km.
The satellite will be deployed during the latter part of flight day two, and
eventually will manoeuvre to a 741 km altitude.
The launch has been scheduled for the night of 15/16 November. The announced
4 h launch period begins at 23:30 UTC on 15 Nov. It has been widely reported
that the actual launch window begins at 23:46 UTC on 15 Nov., and ends at
01:12 UTC on 16 Nov.
The duration of the mission is expected to be four days.
AV WEEK believes that the payload was originally planned to be placed in a
geo-stationary orbit, but was retargetted to provide support for Operation
Desert Shield, in the Persian Gulf.
1.2 (Informed?) Speculation
-----------------------
Some of my friends in the media who cover shuttle missions have expressed
doubts about the AV WEEK story. Their main concern is with the claim that
an imaging reconsat would be placed in a low inclination orbit. There is also
concern about the claim that the payload was originally intended to go into a
geostationary orbit. There are also concerns about the mission being changed
in response to Desert Shield. Here are my views.
1.2.1 Why the Low Inclination Orbit?
------------------------------
At first thought it appears absurd that a U.S. imaging reconsat would be
placed in a low inclination orbit. However, a case can be made for doing so.
In the past, the U.S.S.R. and China were the primary reconnaissance targets,
which necessitated the use of high inclination orbits. Early U.S. reconsats
used 70 deg to 80 deg inclination orbits. Eventually, sun-synchronous orbits,
which have inclinations between about 96 deg and 100 deg (the required
inclination is a function of semi-major axis and eccentricity) became the
standard because they offer near constant sunlight angles from one day to the
next.
The U.S. has three KH-11 (Keyhole) satellites in sun-synchronous orbit:
Int'l NORAD INC PER APO
Name Launch Desig # deg km km
USA 6 KH-11-6 84122A 15423 97.8 335 758
USA 27 KH-11-7 87090A 18441 97.9 291 975
USA 33 KH-11-8 88099A 19625 97.9 292 983
(The orbits of these objects are known through the efforts of amateur
astronomers, who track them.)
USA 27 and USA 33 appear to be fully operational. They use their propulsion
systems to maintain their approximately 300 km by 1000 km orbits. USA 6 was
originally in the same orbit as USA 33, which replaced it. In a break from
past practice, USA 6 was not de-orbited. Since the launch of USA 33, it has
been allowed to slowly decay from orbit, with the exception of a single
manoeuvre in July 1989, which raised its perigee about 50 km, apparently to
reduce drag and prolong the life of the orbit. This suggests that this
satellite may be at least partially operational.
In addition to the KH-11's there are two other satellites which are strongly
believed to be imagers:
USA 34 Lacrosse 88106B 19671 57.0 669 687
USA 53 KH-12 ? 90019B 20516 65.0 806 813
(Our knowledge of these orbits is also due to the efforts of the amateur
astronomers.)
USA 34 was reported by AV WEEK to use a powerful synthetic aperture radar to
enable it to resolve objects to a resolution of about 2 m. It was launched
aboard Atlantis on STS 27.
USA 53 was reported by AV WEEK to be a digital imager. It was launched on
Atlantis on February 28, 1990, and deployed the following day. The Soviets
spotted four large pieces of debris in the payload's orbit on 7 March, and
reported that the satellite had probably "been blown up by its owners". In
October, amateur astronomers found the satellite (by chance) in the orbit
given above.
One feature of USA 53's orbit which tends to support the claim that it is an
imager, is its 9 day (127 rev) repeating groundtrack. Repeating groundtracks
are a common feature of Earth imagers, civilian as well as military.
With four (possibly five) operational imagers in high inclination orbits, and
the greatly improved relations between East and West, it is possible that a
decision was made to place the next digital imager in a low inclination orbit,
to permit better coverage of the Middle East and the drug trade regions.
One argument against this, might be that a high inclination orbit can see
everything that can be seen by a low inclination orbit, so why sacrifice the
high latitudes? Although it is true that a high inclination orbit also
crosses the equatorial region, the spacing between adjacent groundtracks is
farther apart than at the higher latitudes, creating some short term gaps in
coverage. This problem could be reduced through the use of a higher altitude
orbit, at the cost of some loss of resolution, or through the use of a lower
inclination orbit, at the cost of high latitude coverage.
1.2.2 Was the Original Orbit Geosynchronous?
--------------------------------------
It seems highly unlikely that a payload designed for GEO (geosynchronous
orbit), could operate successfully in LEO (low Earth orbit), or vice versa.
For example, a satellite in GEO would use a different system to point its
instruments at targets than one in LEO. Also, there would be a vast
difference in instrument resolution between GEO and LEO.
It is possible that the AV WEEK story confused the terms geosynchronous and
sunsynchronous. Prior to the Challenger accident, the plan was to launch
imaging reconsats into sunsynchronous orbits from the Vandenberg AFB launch
facility. With the post Challenger decision not to launch from Vandenberg,
satellites such as Lacrosse and USA 53 have had to be launched into lower
inclination "compromise" orbits.
1.2.3 Was the Mission Retargetted for Desert Shield?
----------------------------------------------
Planning a shuttle mission requires about 18 months. The planned orbit has a
significant effect on the details of the mission plan. It seems unlikely that
the orbit of this mission could have been changed in less than the three
months between the Iraqi invasion of Kuwait and the publication of the AV WEEK
story.
A more likely explanation is that a decision was made one or two years ago, to
launch this payload into a low inclination orbit, for the reasons given in
Section 1.2.1, contingent upon the success of the USA 53 payload.
1.3 Payload May be Similar to USA 53
--------------------------------
There are some similarities between the STS 36 and STS 38 missions which may
be an indication that the same type of payload is involved in both missions.
Shuttle Atlantis was used for STS 36, and STS 38 will be its first mission
since then. Both missions involve the use of a much lower than normal orbit.
The STS 36 payload, USA 53, repeats its groundtrack every 127 revs (about 9
days). The STS 38 payload's groundtrack will repeat every 128 revs (about 8.8
days).
2.0 Orbital Elements
----------------
The following is a simplified procedure to estimate the orbital elements of
STS 38 and represent them in a pseudo NORAD "2-line" format.
2.1 Inclination
-----------
It is assumed that the inclination will be 28.45 deg as reported by AV WEEK.
2.2 Mean Motion and Rate of Decay
-----------------------------
The 241 km orbit would result in a mean motion of 16.147 revs/day. In case
the orbit turns out to be a little higher or lower, it is recommended that
several mean motions between about 15.9 and 16.3 be used. Orbital decay can
be set to zero because of the uncertainty in the mean motion.
2.3 Eccentricity, Argument of Perigee and Mean Anomaly
--------------------------------------------------
Shuttle orbits are usually close enough to circular that a zero eccentricity
and argument of perigee can be assumed. The mean anomaly will be zero because
the argument of perigee is zero and the epoch will be chosen to coincide with
an ascending node.
2.4 Epoch
-----
The epoch is chosen to be the time of the ascending node (north bound equator
crossing) of the first full revolution of the Earth. For a 28.45 deg
inclination mission, this occurs about 1 h 13 m after liftoff.
The launch time and date must be expressed in UTC (Universal Time). If the
shuttle is launched as expected on 15 Nov at 18:46 EST, then this would be
15 Nov 23:46 UTC. The time of day of the epoch would be :
23:46 UTC
+ 01:13
-----
= 24:59 UTC 15 Nov
= 00:59 UTC 16 Nov
The day of the year is also part of the epoch and is commonly combined with
the time of day of the epoch as follows :
EPOCH = YYDDD.dddddd
where: YY = last 2 digits of year i.e. 90 for 1990
DDD = day of year, i.e. 16 Nov 1990 is day 320
.dddddd = fraction of day, i.e. 00:59 UTC = (0 + 59 / 60) / 24
= 0.040972
Putting the above pieces together yields:
EPOCH = 90320.040972
2.5 Right Ascension of the Ascending Node (RAAN)
--------------------------------------------
The RAAN is a function of the longitude and the time and date of the ascending
node. For the above EPOCH, which corresponds with the ascending node of the
first revolution of a 28.45 deg orbit, the longitude of the ascending node is
-173.2 deg W.
The first step is to calculate the Greenwich mean sidereal time at the epoch.
An accurate formula for 1990 is:
GMST = (6.6265 + 0.06571 * DDD + 24.06571 * 0.dddddd) mod 24
where DDD and 0.dddddd are as defined above
For the epoch calculated earlier the, GMST is :
GMST = (6.6265 + 0.06571 * 320 + 24.06571 * 0.040972) mod 24
= 4.63972 h
The final step is to calculate RAAN :
RAAN = (15 * GMST - WEST LONGITUDE) mod 360
= (15 * 4.63972 - (-173.2)) mod 360
= 242.8 deg
2.6 Summary
-------
The above estimates are summarized below in a pseudo NORAD "2-line" format :
90320.040972 .00000000 00000+00 00000+00
28.4500 242.8000 0000000 000.0000 000.0000 16.147
The first line contains the epoch (Section 2.4) and the three NORAD drag
related quantities, set to zero per Section 2.2. The second line contains
inclination (Section 2.1), RAAN (Section 2.5), eccentricity, argument of
perigee set to zero per Section 2.3, and mean motion (Section 2.2). Remember
to bracket the mean motion between about 15.9 and 16.3.
3.0 Visibility Window Analysis
--------------------------
The tables below show the visibility windows (range of dates of visibility) of
the shuttle during the upcoming mission. There are individual tables for
evening and morning. Visibility windows are a function of time/date of launch
and observer's latitude. The windows have been computed for the start and end
of the expected launch window of 15 Nov 23:46 UTC to 16 Nov 01:12 UTC.
In many cases the windows begin several days prior to the launch date. This
merely indicates when the window would have begun, had the orbit pre-existed
the launch date.
The windows were based on a 241 km, 28.45 deg inclination orbit, as reported
by AV WEEK. If the shuttle goes higher (it can't go much lower), then the
windows generally will be wider. For this project, a window was defined as
passes which culminate at least 5 deg above the horizon, and which are
illuminated for at least half of the pass.
The visibility windows will not be greatly affected by a delay in the date of
launch, as long as the launch window does not change greatly. In case there
is a delay, add the number of days of the delay to the start and end of each
visibility window.
If your latitude is not in the table, then there will be no window. If an
entry for your latitude is blank, then there is no window corresponding to
that launch time.
EVENING VISIBILTY WINDOWS
----------------------------
LAUNCH (UTC) LAUNCH (UTC)
--- ------------- -------------
LAT 15 Nov 23:46 16 Nov 01:12
--- ------------- -------------
40N 14/11 - 18/11
35N 11/11 - 21/11 13/11 - 24/11
30N 09/11 - 23/11 12/11 - 25/11
25N 08/11 - 24/11 10/11 - 26/11
20N 06/11 - 25/11 09/11 - 27/11
15N 18/11 - 26/11 08/11 - 16/11
MORNING VISIBILTY WINDOWS
----------------------------
LAUNCH (UTC) LAUNCH (UTC)
--- ------------- -------------
LAT 15 Nov 23:46 16 Nov 01:12
--- ------------- -------------
10S 10/11 - 16/11
15S 09/11 - 16/11 12/11 - 19/11
20S 10/11 - 27/11 13/11 - 30/11
25S 12/11 - 26/11 14/11 - 29/11
30S 14/11 - 25/11 16/11 - 28/11
35S 16/11 - 24/11 18/11 - 26/11
4.0 Observation Tips
----------------
The shuttle is easy to spot with the naked eye. When favourably illuminated,
nearly overhead and in a dark sky, it has a visual magnitude between -1 and
-2, about as bright as Jupiter. The shuttle has been observed as early as
15 minutes after sunset or before sunrise, however that is probably too
difficult for the inexperienced observer.
The uncertainty in the mean motion makes the search for the shuttle a
challenge, but far from impossible. The best search strategy is to produce
several different orbital element sets covering the range of uncertainty in
the mean motion, as recommended in Section 2.2. In this way the predictions
will "bracket" the shuttle's actual time of passage and path across the sky.
This procedure takes advantage of the fact that the orientation of the
shuttle's orbital plane with respect to the Earth can be predicted with much
greater accuracy than the position of the shuttle within its orbit. The idea
is to "stare" at the imaginary ring in the sky which is the shuttle's orbit.
As you wait for the shuttle to appear, the Earth rotates, which makes the
orbit ring move across the sky. The shuttle must occupy each point along the
orbit once per revolution, so eventually it must be seen.
If the shuttle makes a near overhead pass, even the small uncertainty in the
orientation of the plane can result in large errors in its predicted path
across the sky, especially at maximum elevation. Therefore, take care to scan
a wide section of the sky. It would be unfortunate to be looking for a 65
degree high pass in the south only to have the shuttle pass behind your back,
70 degrees high in the north.
5.0 Observation Network
-------------------
During the STS 27, STS 28, STS 33 and STS 36 DoD missions there was an
informal network of amateur observers who found the shuttle and shared their
observations. This made it possible for more people to see the shuttle
because we were able to quickly refine our orbital elements and pass on the
information. The following sections discuss making orbital predictions,
accurate observations and how to report your observations.
5.1 Making Predictions
------------------
You require an orbit prediction program to make use of the elements from
Section 2. SEESAT, by Paul Hirose is a good public domain program for this
purpose. It enables you to make accurate predictions for any location on
Earth. Predicted positions are given in azimuth and elevation as well as
right ascension and declination. The program was written in C and has been
compiled for the IBM-PC. The source code has been included to enable use on
non-IBM systems. You can download a copy from the Canadian Space Society BBS,
at the number given in Section 5.3. The name of the file is SEESAT2X.ARC.
If you are not into making predictions but wish to make observations, I will
try to provide you with predictions. Contact me via one of the channels listed
in Section 5.3, well in advance of the launch.
5.2 Making Accurate Observations
----------------------------
The best observations are positions related to the stars along with the time
accurate to 1 second or better. For example,
"passed between Castor and Pollux, 1/3 distance from Castor to
Pollux, 08:34:21 UTC 22 Feb 1990"
or
"passed 3 degrees below Vega, 09:12:10 UTC 22 Feb 1990"
In addition, estimates of visual magnitude and colour would be useful. If the
magnitude is varying regularly, measure the period of variation. If two
objects are seen, then state the separation between them. For example, "the
brighter object led the fainter by 10 seconds of time", or "the red object
was about 4 degrees behind the other at maximum elevation of 50 degrees" would
be useful.
Also, if possible, note the time and approximate position of shadow entry or
exit. This can be very helpful in determining the orbital eccentricity and
argument of perigee.
Make certain to provide your latitude and longitude as accurately as possible.
5.3 Sharing Your Observations
-------------------------
If you have information to share, use one of the following communications
channels. I will attempt to respond in kind. If you make an observation on
the first day of the mission, please if possible, phone it in to me. The
initial observations are very important in determing an accurate orbit so that
a maximum number of people will have an opportunity to observe later orbits.
1) Leave a message on the CSS (Canadian Space Society) BBS
for Ted Molczan. This is a free, 24 h/d board, 2400 8N1,
(416) 458-5907. This will be the primary communications
channel.
2) Phone me at (416) 921-1564
3) Send e-mail message to molczan@gpu.utcs.utoronto.ca
This is on NETNORTH, which is connected to BITNET and
other academic/research networks.
Please pass this on to other BBS's or interested individuals.
* * * *
--
Ted Molczan@gpu.utcs.utoronto.ca