Thursday, June 30, 2016

USA 224 recovered: an update of the KH-11 constellation

Russell Eberst in Scotland has recovered the noon-plane KH-11 ADVANCED CRYSTAL/"Keyhole" optical reconnaissance satellite USA 224 (2011-002A) this week. The recovery happened relatively late (in 2015 it was recovered 2 months earlier).

This recovery means that, after the preliminary update last March, I can provide my periodic update on the orbits of the KH-11 constellation based on timely orbital data.

In various previous post to this blog, I outlined how the KH-11 constellation consists of two primary orbital planes, the primary East and West planes; and originally two, now one, secondary orbital plane(s). Of the latter secondary planes, only one, the secondary West plane, is left after the de-orbit of USA 161 late 2014.

The past decade or so, the primary planes have been 48-49 degrees apart in RAAN. That is still the case: USA 224 and USA 245, the primary East and West plane KH-11's, are currently 49 degrees apart in RAAN.

The secondary planes used to be either 10 or 20 degrees from the corresponding primary plane in RAAN, but since mid-2014 the secondary West plane (currently USA 186) has moved further out, to 24 degrees West of the primary orbital plane.

As I have outlined before on this blog, the secondary plane(s) differ in orbital altitudes from the primary planes. The current configuration:

         perigee   apogee    l time   repeat
Sat        km        km      d node   (days)   plane
USA 186    261       454     08:05      3      secondary W
USA 224    262      1007     12:58      4      primary E
USA 245    266      1000     09:42      4      primary W

Given are the apogee and perigee altitudes of the satellites, the average local time they pass through their descending node (an indication of around what time they pass a given area - all satellites in the constellation are sun-synchronous, i.e. they pass  at a similar solar elevation each day), the repeat interval of the ground track in days, and the plane they orbit in.

What can be seen is that the secondary plane satellite, USA 186, is in a much more circular orbit with a much lower apogee (454 km), compared to the two primary satellites (~1000 km). Perigee altitudes of all three satellites are similar. I have speculated on the reason for this apogee difference of the secondary plane satellite at the end of a previous post.

The West plane satellites, USA 186 and USA 245, make morning passes, about 1h45m after each other. The East plane satellite, USA 224, makes passes about an hour after local noon.

The current orbital configuration has been more or less stable since mid-2014 (or more exactly, since USA 161 was de-orbitted late 2014).

Labels: , , , , , ,

Monday, June 27, 2016

Mentor 7 (NROL-37) stopped drifting at 102.6 E

Mentor 7 on 25 June 2016 
image (c) Paul Camilleri, used with permission
click to enlarge

On June 11, 2016, the National Reconnaisance Office (NRO) launched NROL-37: a new Mentor (Advanced ORION) SIGINT satellite, Mentor 7 (2016-036A). Paul Camilleri in Warners Bay, Australia, located it in orbit three days later, on June 14 (see a previous post).

At that time, it was in a semi-geosynchonous, 7.5 degree inclined drift orbit, and drifting westwards in longitude at a rate of ~0.28 degrees/day (see a previous post), after initial orbit insertion near longitude~105 E.

New observations by Paul Camilleri on June 24 and 25 show that this drift has stopped. The satellite is now geosynchronous in a stable, 7.5 degree inclined position at longitude 102.6 E. It arrived there on June 19th, after a 7-day drift.

click map to enlarge

This is almost certainly a temporary check-out position. In this location the satellite is positioned at 45 degrees elevation (i.e. halfway between zenith and horizon) for the Pine Gap Joint Defense Facility in central Australia, one of the primary ground stations for US SIGINT satellites:

Mentor 7: position as seen from Pine Gap
click to enlarge

It will probably remain here for a few weeks or a few months, and then be moved to an operational location, which I suspect will be near longitude 80 E.

Current elements:

Mentor 7
1 41584U 16036A   16177.93784503 0.00000000  00000-0  00000+0 0    01
2 41584   7.5070 353.7330 0045273  39.1128 322.1888  1.00270000    04

Labels: , , , ,

Friday, June 24, 2016

MUOS 5 GTO insertion and Centaur fuel dump imaged from Australia

click to enlarge
image (c) Paul Camilleri - used with permission

The spectacular image above was kindly made available to me by Paul Camilleri from Warners Bay in Australia. Taken around 18:03 UT using a 180 mm lens, it shows the just launched MUOS 5 satellite and the associated Centaur upper stage: the latter is venting fuel creating a "comet-like" cloud.

The image was made some 40 minutes after MUOS 5 separated from the Centaur stage (separation happened at ~17:23 UT). The two objects were at an altitude of ~30 000 km at that time, in a Geosynchronous Transfer Orbit (GTO).

Following separation, the Centaur upper stage had made a Collision and Contamination Avoidance Manoeuvre (CCAM) and next started to dump exces fuel in order to reduce the risk of later on-orbit explosions. This fuel-venting causes the comet-like cloud. MUOS 5 itself is visible as a small trail just under the Centaur and its fuel cloud.

Two other classified objects are, by chance, visible in the image as well: Milstar 4 and USA 155. Like MUOS 5, Milstar 4 is a military communications satellite: USA 155 is an SDS data relay satellite.

MUOS 5 was launched today at 14:30 UT (24 June 2016) from Cape Canaveral, using an Atlas V rocket with a Centaur upper stage. For a timeline and details, see here.

Over the next couple of days, MUOS 5 will use its own engines to make a series of orbit raising manoeuvres, followed by an orbit circularization to bring it in a ~5-degree inclined Geosynchronous orbit. Most likely it will initially be placed in a check-out position near longitude 172 W: I observed MUOS 4 in this position last year.

After 5 months or so, when check-out is completed, it will next be moved to longitude 72 E, where it will be parked as an on-orbit spare in the MUOS constellation (see also my earlier post on MUOS 4 here).

MUOS 5 is the fifth satellite in the Mobile User Objective System (MUOS) system of Geosynchronous narrowband communication satellites. The first MUOS satellite was launched in 2012. This system of military COMSAT is to provide communication facilities to 'mobile users': i.e. military personel in non-fixed positions such as ships, aircraft, tanks and vehicles or on foot. It is a replacement for the aging UFO constellation of COMSAT and will be able to be used by legacy UFO equipment.

The MUOS system now consists of four operational satellites (MUOS 1 to 4) and MUOS 5 as said is to function as an on-orbit spare. According to a publication by Oeting et al. in the Johns Hopkins APL Technical Digest 30:2 of 2011, it will be parked at 72 E for this purpose.

I thank Paul Camilleri for permission to feature his splendid image!

Labels: , , ,

Sunday, June 19, 2016

Updated orbit for Mentor 7 (NROL-37 payload)

In my previous post I reported that the geosynchronous payload of June 11th's NROL-37 launch, the SIGINT satellite Mentor 7 (USA 268, 2016-036A) was found on June 14 by Paul Camilleri in Australia.

Paul has communicated new observations from June 15 and 16, extending the observational arc to 2.1 days. I fit the following updated orbit to it:

Mentor 7
1 41584U 16036A   16167.96105997 0.00000000  00000-0  00000+0 0    07
2 41584   7.5055 353.7008 0046333  41.2140 319.1375  1.00195548    05

rms 0.004 deg      from 9 obs June 14.70 - June 16.79  (2.09 day arc)

This orbit results in a drift rate of ~0.28 degrees per day in longitude, westwards. If this drift rate does not change in the future, the satellite will reach longitude 80 E (my guess for its eventual operational position) at the end of the first week of September 2016 [update 27 June: but see follow-on post here].

More on Mentor 7 and its recovery (including one of Paul's recovery images) in my previous post.

UPDATE 27 June 2016: Mentor 7 has stopped drifting and is stable at longitude 102.6 E - more on that in this follow-on post.

Labels: , , , ,

Saturday, June 18, 2016

Mentor 7, the NROL-37 payload, found

Launch of NROL-37 (photo credit: ULA)

On 11 June 2016 at 17:51 UT, after a one-day postponement, the US National Reconnaissance Office (NRO) launched a classified payload from Cape Canaveral under the launch designation NROL-37. It was a launch into geosynchronous orbit using a Delta IV-Heavy rocket.

The NROL-37 payload  has been catalogued under the generic designation USA 268 (2016-036A, 41584). It is widely believed to be a Mentor (Advanced Orion) SIGINT ('eavesdropping') satellite, Mentor 7.

Initial assessments pre-launch indicated a possible orbit insertion of the payload over Southeast Asia. After launch, Paul Camilleri, a novice satellite observer in Australia, was guided by Ted Molczan and me in an attempt to find the payload by means of a dedicated photographic survey.

In the early morning of June 15 (local time -  June 14 in UT), three days after the launch, Paul indeed successfully located the payload! The image below shows one of Paul's initial images, with the NROL-37 payload visible as a bright dot.

Mentor 7 (NROL-37) imaged June 14 by Paul Camilleri in Australia
click to enlarge - photo (c) Paul Camilleri, used with permission

From imagery on June 14 and 15, the following very preliminary orbit was calculated (for the time being, I have fixed a few parameters towards 'round' values here):

Mentor 7
1 41584U 16036A   16166.96303997 0.00000000  00000-0  00000+0 0    06
2 41584   7.5000 353.7000 0046000  41.4155 318.9349  1.00200000    04

rms 0.006, from 7 obs, 2016 June 14.70 - June 15.48 UTC

This places the satellite near longitude 104 E, over the Strait of Malacca, around the time of discovery, in a ~7.5 degree inclined near-geosynchronous orbit.

[edit 19 June 2016, 20:15 UT: I have posted an updated orbit in a later post here]

click map to enlarge

While the Mean Motion still remains somewhat ill defined from this short an observational arc, the satellite appears to be slowly drifting westwards, towards its eventual operational position.  My guess (and no more than that) is that it will eventually stop drifting near either 80 E (south of Sri Lanka) or perhaps 10 E (over central Africa). The reason for the initial placement near 104 E is likely that in this position it is initially well placed for the Pine Gap Joint Defense Facility ground station in central Australia (one of two facilities dedicated to NRO SIGINT payloads) during the initial check-out phase.

Mentor (Advanced Orion) satellites are SIGINT satellites: satellites that "listen" for radio signals. They are "the largest satellite[s] in the World", according to a statement by the then NRO director Bruce Carlson in 2010 at the time of the Mentor 5 (NROL-32) launch. There has been some speculation (it seems to be not more than that) that these satellites might have a huge fold-out mesh antenna some 100 meters wide.

Our observations suggest that these satellites indeed appear to be extraordinarily large. They are very bright (brighter than other geosynchronous payloads), typically of magnitude +8. They are the easiest geosynchronous satellites to photograph: a standard 50mm lens with a 10-second exposure will do.

The other six Mentor satellites, launched between 1995 and 2012, currently make up this configuration:

click map to enlarge

I thank Paul Camilleri for permission to use one of his photographs and for his willingnes to undertake the hunt for Mentor 7

 [edit 19 June 2016, 20:15 UT: an update here]

Labels: , , , ,

Tuesday, May 24, 2016

Geostationary bonanza

click to enlarge

The images above and below are two small parts of one single image shot on May 2nd 2016, using a SamYang 1.4/85mm lens on a Canon EOS 60D with 30 seconds exposure (ISO 1000) under a very dark sky. These two image excerpts overlap in the corner: the upper right corner of the image excerpt above overlaps with the lower left corner of the image excerpt below.

Although both sub-images are only a few degrees wide, they show a bonanza of objects, including 3 classified objects.

In the image above, 11 objects including the classified SIGINT satellite PAN (2009-047A) are visible. PAN is parked next to the commercial communications satellite Yahsat 1B.

In the image below, 10 objects including two classified objects are visible: the two classified objects are the SIGINT satellite Mercury 1 (1994-054A), and the SIGINT satellite Mentor 4 (2009-001A), the latter parked next to the commercial communications satellite Thuraya 2.

The full 10 x 14 degree image, of which the images featured here are small excerpt parts, shows over 30 objects.

click to enlarge

Labels: , , ,

Sunday, April 03, 2016

More on the Hitomi/ASTRO-H breakup

Hitomi and its debris train
(click image to enlarge)

In a previous post from a few days ago, I discussed the debris shedding event of the Japanese scientific Hitomi/ASTRO-H satellite (2016-012A) on March 26, ~01:42 UT, that is currently in the news. As things have developed, it appears that the picture has become more grimm, as the available data now seem to indicate a real break-up rather than just "some" debris shedding. If true, then Hitomi is beyond saving.

On Friday, JSpOC published the first elements for 10 fragments produced. As part of it, the "A" identity (the largest remaining part of the original satellite body) was shifted to another object: what was assigned to be the remnant of Hitomi in the first few days, turns out to be not, but was the fragment now labelled "L". It was this fragment which was observed tumbling with a periodicity of 11.75 seconds. As it is rather bright, it indicates that Hitomi broke into at least two large pieces (I think it is actually three, as fragment H also seems to be large based on the NDOT/2 value) plus a lot of smaller fragments.

Based on the published elements, I calculated the following approximate delta V's for these fragments based on change in semi-major axis and inclination (velocity in meters per second):

fragment  number    delta V

A         41337     0.4 m/s
N         41444     0.3 m/s
Q         41446     0.4 m/s
P         41445     0.5 m/s
R         41447     0.5 m/s
L         41442     1.3 m/s
H         41439     1.4 m/s
J         41440     3.9 m/s
G         41438     7.2 m/s
M         41443    12.2 m/s
K         41441    13.7 m/s 

Some clustering might be present in these speed data, roughly into three groups: a number of fragments yield low velocities in the range 0.3-0.5 m/s. Fragments L and H have ~1.3-1.4 m/s. And there is a group with high speeds from ~4 to ~14 m/s.

I performed a conjunction analysis using the first two to three published element sets for each fragment, and Rob Matson's COLA software. For a given fragment, I looked for the element sets with small miss distances and relatively low ephemerid ages relative to the calculated conjunction.

The result is in the diagram below. Note that some fragments appear twice in the diagram, as they sometimes yield two equally well fitting close approach times. There clearly is scatter and ambiguity in the results.

The data nevertheless show a clear cluster close to the estimated time that JSpOC published, 01:42 UT on March 26. The strongly deviating times to the right in the diagram are probably inaccurate and concern small fragments with a quick orbit decay (large NDOT/2 value), the reason why they are less accurate.

click diagram to enlarge

I find it interesting that the A fragment seems to plot somewhat earlier than all the others, near 01:10 UT, but think it is wisest to chalk this down to orbit inaccuracies.

Ted Molczan has argued for a possible second debris-shedding event at ~02:35 UT, near the next perigee passage. Fragment K, but in my opinion perhaps also P and Q would fit this time (P however also loosely fits to the 01:42 UT cluster. The Q fragment is probably small and has a larger miss-distance in the analysis, so might not be too accurate).

In order to understand what happened to Hitomi/ASTRO-H, a so called "Gabbard diagram" can be helpful. In a Gabbard diagram, the orbital period of each fragment is plotted against the apogee and perigee altitudes. That results in this diagram:

click diagram to enlarge

The relative position in the diagram with respect to the original values for Hitomi/ASTRO-H before the event happened (indicated by solid dots, one for apogee and one for perigee), is informative.

Fragments plotting to the right and up of the original position are fragments that were ejected into the direction of the original satellite movement (prograde). They get longer periods and somewhat higher orbital altitudes. Fragments plotting to the left in the diagram, are fragments ejected opposite to the movement of the satellite in its orbit (retrograde). Fragments ejected perpendicular to the plane of movement of the satellite will plot in the vicinity of the original satellite values.

What can be seen in the diagram is that what is presumably the largest remaining object, fragment A, plots just right of the original Hitomi position: orbital period and apogee altitude slightly increased. A few more object plot around the original Hitomi/ASTRO-H values, but most of the fragments plot to the left: their orbital periods and orbital altitudes decreased. This includes the L-fragment, which based on visual observations and the NDOT/2 value relative to other fragments is likely a large fragment, as is probably fragment H.

I interpret this as follows: as indicating breakup from an origin somewhat behind the center of mass of the satellite (with respect to its direction of movement). This gives the heaviest remaining body (the A fragment), predominantly material originally located near/in front of the center of mass, a momentum in the direction of movement. Most other, smaller parts appear to have been predominantly ejected backwards, which is perhaps some indication that predominantly the 'rear' part of the satellite exploded with a notably backwards impulse.

This still says little about the cause of the fragmentation. My primary suspicion is however that either the liquid Helium tank of the Soft X-ray Spectrometer (SXS) situated near the base plate in the middle of the spacecraft exploded, or the nitrogen pressure tank of the thruster system (I am not sure where the latter is situated in the satellite, I have not been able to trace that information so far).

A preliminary analysis suggests that objects G and M will re-enter the Earth's atmosphere somewhere mid-April.


[update 4 April 2016, 9:00 UT] My (and Google translate's) Japanese is abysmall, but this Japanese article seems to suggest the Helium tank indeed might be to blame, if I get it right. It also seems to suggest there was an attitude anomaly half a day before the main break-up, around 13:10 UT on March 25th.

[update 5 April 2016 7:30 UT] Thanks to a translation by my friend Ton Kruijer and a Twitter comment by Nobuyuki Kawai: what the Japanese article from the previous update says is that JAXA rules out an impact as the cause and thinks something happened "internally", in the spacecraft. No definite cause is identified, but the article speculates about the battery and Helium tank used for cooling SXS. There are however no clear indications about the cause in the telemetry received before contact was lost.

Labels: , , ,

Thursday, March 31, 2016

USA 186 rising, and revisiting the KH-11 constellation

click image to enlarge

The image above shows the KH-11 ADVANCED CRYSTAL/"KeyHole" optical reconnaissance satellite USA 186 (2005-042A) rising over the roof of my house in late twilight of March 28, 2016.

It had passed perigee at 260 km altitude just 10 minutes earlier and was at about 285 km altitude when I photographed it. At that low an altitude, it races over the sky (these are 5 second images with a 50mm lens). Clearly visible by the naked eye, I watched it rise over the roof, and then slowly flare to mag. -1 near 19:17:55 UT:

USA 128 186 flaring near 19:17:55 UT (March 28, 2016)
click image to enlarge

USA 186 is the oldest of the remaining KH-11 ADVANCED CRYSTAL satellites. After the de-orbit of  USA 161 late 2014, there are now three KH-11's left on-orbit:  

(a)  USA 186  (2005-042A) pictured here;
(b)  USA 224  (2011-002A), and;
(c)  USA 245  (2013-043A).

USA 224, the primary East plane satellite, and USA 245, the primary West plane satellite, are separated by 48.7 degrees in RAAN. USA 186, the secondary West plane satellite, is 24.3 degrees (half the separation of the two primary satellites) West of USA 245 in RAAN:

click to enlarge

All the orbits are sun-synchronous, meaning that they maintain an orbital precession coupled to the daily precession of the sun.This ensures that they make passes at roughly the same time each day, imaging terrain with roughly similar solar elevations. This helps interpret imagery, as the shadow patterns will be similar on images from nearby days and changes in shadow pattern will stand out.  

USA 224, the primary East plane satellite in the constellation, makes daily passes around 13h local solar time. It's ground track repeats each 4 days. USA 245, the primary West plane satellite, makes daily passes around 9:45h local solar time, also with a ground track repeating each 4 days. USA 186, the secondary west plane satellite, makes daily passes around 8h local time, with a repeating ground track each 10 days.

Basic data ():

sat      orbit           mlt DN    repeat  plane 

USA 124  265 x 1009 km*  13:00     4 days  primary E
USA 245  268 x 1020 km   09:45     4 days  primary W
USA 186  260 x 456  km   08:05    10 days  secondary W

* September 2015 (sat not yet recovered in 2016)

("mlt DN" means mean "local time of passage through Descending Node") 

Note that due to the winter blackout in the Northern hemisphere we haven't observed USA 224 for seven months: the orbit shown here assumes it has kept up its September 2015 orbit. We will know for sure when we recover the object in early summer.

The previous secondary West plane satellite in the constellation, USA 129 (de-orbited in late April of 2014) used to make passes near 9h local time. By changing the RAAN difference between the primary and secondary satellite from 10 degrees to 24.3 degrees after USA 186 moved from the primary to the secondary plane and took over from USA 129 following the launch of USA 245 (see earlier post here and links theirin), the pass time was shifted to one that occurs one hour earlier.

As can be seen in the table above, all current KH-11's have perigee at a roughly similar altitude of ~260-270 km. The primary plane satellites have apogee at ~1010-1020 km. The secondary plane satellite has a much lower apogee, at ~455 km altitude.

The exact reason why the secondary plane KH-11's lower their apogee is unclear. USA 129 was the first to lower apogee connected to its move from the primary West plane to the secondary West plane in 2006. USA 161 repeated this pattern in 2011 when it was moved from the primary East plane to the secondary East plane (since the de-orbit of USA 161 late 2014, this latter secondary East plane no longer exists). Various options can be considered:

(a) some operational constraint, e.g. aim for a particular ground-track repeat interval;

(b) some operational constraint, e.g. related to image resolution (this is unlikely);

(c) some operational constraint, e.g.different imaging role compared to the primary plane satellites;

(d) making it easier to de-orbit the satellite near end-of-life: if fuel reserves are low, a lower apogee would allow de-orbit from apogee if necessary, so one does not necessarily have to wait for perigee to be located over the southern Pacific.

Labels: , , , ,

Tuesday, March 29, 2016

Hitomi (ASTRO-H) scientific X-ray satellite suffered an orbit-altering event

On 26 March 2016 near 1:42 UT, the Japanese Hitomi/ASTRO-H satellite, a scientific Röntgen space telescope launched February 17, suffered a clearly energetic event that changed its orbit and resulted in a communications loss. JSpOC reported the release of at least 5 fragments from the spacecraft. The existence of these fragments was first detected by US tracking stations near 8:20 UT. Observations by amateur satellite observers reported on Seesat-L (here, here, here) suggest that the spacecraft is tumbling with a periodicity of about 23.5 seconds (flashes each ~11.75 seconds) according to a preliminary analysis by Ted Molczan here.

click to enlarge map

Analysis of the pre- and post-event orbital elements suggests that the event occurred near 1:40-1:50 UT on March 26 (see note at end of post (*). JSpOC estimates the event occured at 1:42 +/- 11 m UT. The satellite was just past perigee and had just passed the ascending node of its orbit at that moment, moving over Meso-America (see map above). The nominal time for the event, 1:42 UT, is about 12 minutes past nodal crossing.

The Japanese Space Agency JAXA reports that since noting loss of communications at 7:40 UT on March 26, they have managed to receive signals from the spacecraft twice, near 13h and 15:30 UT on March 28.

The event was energetic enough to alter the orbit of the satellite, slightly increasing its eccentricity and dropping slightly in altitude: it's semi-major axis changed by 2 km, with perigee dropping by 4.5 km. The inclination was changed too, by about 0.0045 degrees.

(click diagrams to enlarge)
data source: Space-Track

Based on the change in semi-major axis and inclination, and if my calculations are correct, the orbit change of ~2 km in semi-major axis and ~0.0045 degree in inclination is the result of a velocity impulse with delta V of about 1.2 m/s. The most likely source is a sudden venting or a small explosion of some sort. An explosion of some sort also explains the discharge of five fragments reported.

The time of the disruptive event is almost exactly 6 hours before the satellite was to turn operational.

How serious the event is, is unclear at the moment. The fact that the spacecraft still appears to be sending signals is a positive aspect. If JAXA can re-establish contact and stop the tumbling, perhaps the mission can be saved. Or perhaps not, depending on the damage to the spacecraft.

* Note: my initial analysis of pre- and post event orbital elements on March 28 yielded times near 1:40 UT (March 26), close to JSpOC's nominal 1:42 UT time for the event. Annoyingly, I had some odd brain malfunction next and tweeted "4:40 UT" instead. 

It was probably induced by a mix-up in my mind of the established 1:40 UT event time with the epoch time of the first abberative TLE, which was16086.196059 = 26 Mar 4:42 UT. I had jotted both times down and then mixed them up apparently...such things happen when you spent too much time coding and processing data in an excel spreadsheet.

The way in which I analyzed the probable time of the event, is by taking pairs of pre- and post-event elsets for the satellite and treating them as if they concerned two separate objects. I then used Rob Matson's COLA software to establish close encounters between the "two" objects: near the time of the event, the orbital positions from both elsets should nearly coincide. Depending on which elsets you pair, this yields times scattering between 1:30 and 2:06 UT . My initial test of four elset pairs had three of them yielding times near ~1:40 UT.

Labels: ,

Saturday, March 12, 2016

MUOS 4 recovered at 75 E

MUOS 4 at 74.8 E on 8 March 2016
image (c) by Greg Roberts, South Africa
(click to enlarge)

In a recent post I wrote that MUOS 4 (2015-044A) had left it's check-out position at 172 W near Hawaii late January, drifting westwards. I presumed that it was being moved to its assigned operational spot at 75 E, over the Indian Ocean just south of India.

This is now confirmed. On 8 March 2016, Greg Roberts from Cape Town, South Africa, recovered MUOS 4 at 74.8 E. Greg's recovery image (used with his permission) is above. As I wrote before, it probably arrived there on 29 February 2016, after a 37-day drift westwards at a rate of ~3 degrees/day.

With four satellites at their operational positions, the MUOS constellation is now complete.

MUOS 1   2012-09A       177 W  Pacific
MUOS 2   2013-036A      100 W  CONUS
MUOS 3   2015-002A     15.8 W  Atlantic

MUOS 4   2015-044A       75 E  Indian

However, one more MUOS satellite will be launched. This fifth satellite will be parked at 72 E and will function as an on-orbit spare, in case one of the other four MUOS satellites malfunctions on-orbit.

(click to enlarge)

As can be seen in the illustrations above, the MUOS satellites are separated by ~90 degrees in longitude, but with a slightly bigger gap (~108 degrees) between MUOS 1 and MUOS 4, a gap representing the Pacific. The latter is probably in order to assure access to/from at least two ground facilities, with Hawaii and California serving MUOS 1. The latter would not have been possible with MUOS 1 at ~90 rather than 108 degrees from MUOS 4. MUOS ground facilities are indicated by yellow squares in the map above.

Labels: ,

Friday, March 11, 2016

Imaging a "UFO" (Ultra High Frequency Follow-On)

UFO F2 on 3 March 2016
(click image to enlarge)

The image above is my first image of a UFO...

(* cue X-Files tune *)

No need to call in Mulder, however. The object in the image is a geosynchronous satellite, UFO F2 (1993-056A).

The truth is out there

The acronym 'UFO' in this case does not stand for the classic Unidentified Flying Object. It stands for Ultra High Frequency (UHF) Follow-On, the name of a class of US Navy communication satellites.

The UFO satellite constellation consists of 11 satellites (not all of them operational) in geosynchronous orbit, launched between 1993 and 2003. It serves fleet-wide communication needs for the US Navy (including its submarines, but also Marine units on land). The system is currently being replaced by the newer MUOS constellation (see a previous post) and will gradually be phased out.

UFO satellite constellation on 9 March 2016
(click image to enlarge)

The first launch in the series, the launch of UFO F1 on 25 March 1993 with an Atlas 1 from Cape Canaveral, resulted in a partial failure to reach the intended geosynchronous orbit due to the failure of one of the rocket engines. The second UFO launch, UFO F2, the one imaged above, was the first truely successful launch of this satellite class.

USA 236 on 28 February 2016
(click image to enlarge)

I imaged more geosynchronous objects the past week, taking advantage of clear moonless evenings. The image above shows a star field in Orion in the evening of 28 February 2016, with USA 236 (SDS 3 F7, 2012-033A), an SDS data communications satellite in geosynchronous orbit. These satellites relay data from other US military satellites, optical and radar reconnaissance satellites in Low Earth Orbits such as the KH-11 'Keyhole'/CRYSTAL, Lacrosse (ONYX) and FIA (TOPAZ), to the US.

PAN on 28 February 2016
(click image to enlarge)

I also did my periodic revisit of the enigmatic SIGINT satellite PAN (2009-047A) as well (see image above). PAN is still stable at 47.7 E (see my long-term analysis here), near Yahsat 1B. The image above shows it near that satellite and a number of other commercial communications satellites in an image taken on 28 February 2016.

Mercury 1 r on 3 March 2016
(click image to enlarge)

On Feb 28 and March 3, I recovered Mercury 1 r (1994-054B), the upper stage from the launch of the Mercury 1 SIGINT satellite. We had lost this object for a while, it had not been seen for 153 days when I recovered it. The image above shows it in Hydra on 3 March 2016.

USA 186 on 5 March 2016
(click image to enlarge)

As spring is approaching, the visibility of satellites in Low Earth Orbit is gradually coming back for northern hemisphere observers.  This means we can take over from our lone southern hemisphere observer, Greg. The image above shows the KH-11 'Keyhole'/CRYSTAL optical reconnaissance satellite USA 186 (2005-042A) imaged on 5 March 2016.

Labels: , , , , , , , , , ,

Sunday, March 06, 2016

The tumble period of the UNHA-3 upper stage from the recent North Korean launch is slowly changing

click image to enlarge

The image above, taken in the evening of 5 March 2016,  is a 10-second exposure showing several flashes of the tumbling UNHA-3/Kwangmyŏngsŏng rb 2016-009B, the upper stage from North Korea's recent Kwangmyŏngsŏng-4 launch. It was taken during a very favourable 67-degree elevation pass, using my Canon EOS 60D and a SamYang 1.4/85mm lens (set at F2.0). The sky had cleared just in time for this pass (a last wisp of clouds is still visible in the image).

The flashes had a brightness of about mag. +3.5 and were visible by the naked eye. The resulting brightness variation curve is this one:

click diagram to enlarge

I have briefly mentioned the tumbling behaviour of this rocket stage in an earlier post. Over the past week I have been following this rocket when weather allowed, obtaining observations in the evenings of Feb 28, Feb 29, March  3 and March 5. This now allows a first look at how the tumble rate is (very) slowly changing.

The theory behind tumbling rocket stages and why their tumble rate changes over time, is briefly discussed here on the site. After the payload and the upper stage separate, usually by means of exploding bolts, the upper stage gets a momentum from this separation.

Over time, the resulting tumble is influenced by interaction of the rocket stage body with the earth's magnetic field. Spent upper stages are basically hollow metal tubes, and the Earth's magnetic field causes induction in it, leading to the tube getting an electric charge. Basically, the rocket stage becomes a dynamo. The Earth's magnetic field then further interacts with this electrically charged rocket stage, by means of the Lorentz force exerting a magnetic torque on the rocket stage's spinning motion. It is the latter effect which by "tugging" on the tumbling stage, changes its momentum, with a changing tumble period as a result. The resulting change is one towards a slower tumble rate, and eventually the stage might stop tumbling altogether.

I earlier established a peak-to-peak period of 2.39 seconds for 2009-009B from observations on Feb 28 and 29. Analysis of the new data obtained on March 3 and 5 show that the period is changing: I get 2.43 seconds for March 3 and 2.45 seconds for March 5.

I re-analyzed the Feb 28 and 29 data as well, this time using a fit to a running 5-point average on the raw data, which leads to somewhat better refined peaks. I also found that the initial autofit made by PAST is actually not the best fit, based on the r-square values of the fit. re-analysis leads to a 0.01 second revision to 2.38 seconds of the Feb 28 period, while the Feb 29 period stays at 2.39 seconds as initially established.

So the sequence is:

peak-to-peak periods
Date        TLE date    Period(sec)
Feb 28.81   16059.81    2.38 ± 0.01
Feb 29.79   16060.79    2.39 ± 0.01
Mar 03.79   16063.79    2.43 ± 0.01
Mar 05.82   16065.82    2.45 ± 0.01

(NB: the listed uncertainty is an estimate)

Even though the differences are very small, there appears to be an increasing trend to the periodicity, at the rate of about 0.01 second per day. As the difference is systematic, it is probably real and not just scatter due to measuring uncertainty (time will tell if this indeed holds).

[edit 7 March 2016 19:55]
One caveat: the synodic effect. As the viewing angle changes over the pass, this has some influence on the determined period. For fast tumblers this effect is small, but as we are talking about differences in the order of a few 0.01 seconds, the synodic effect comes into play.
The observations of Feb 28, 29 and March 3 were all made some 30 degrees beyond culmination, so the synodic effect should be about the same. The March 5 observation was done at culmination (I actually have a second image post-culmination as well but have not analysed it yet)
[end of edit]

Below are the brightness curves on which these values are based (click diagrams to enlarge):

click diagrams to enlarge

Appendix: on the construction of these brightness curves

I got a number of questions on how, and with what software, I produce these brightness curves. I will briefly explain below.

(a) calibrate exposure duration
What is first necessary, is that the real duration of the exposure is carefully calibrated. A "10-second" exposure set on your camera is not exactly 10.000 seconds: with my Canon EOS 60D for example, it is 10.05 seconds in reality (this deviation seems to increase exponentially with exposure time: a "15-second" exposure for example is in reality closer to 16 seconds!).

(b) measure pixel values with IRIS
The pixel brightness over the trail on the photograph is measured using the free astrophoto software IRIS.  Load the image, and chose "slice" from the menu option "view". Put the cursor at the start of the trail, and draw a line over the trail to the end of the trail. A window pops up with a diagram. You can save the data behind this diagram as a .txt table.
NB: be aware that Iris always measures from left to right (no matter how you draw the line), so if the satellite moved from right to left, you will later have to invert the obtained data series.

(c) Excel manipulation
The resulting .txt data file is read into excel. There, if necessary I first invert the series (see remark above). The result is a table with a column with pixel brightness values,  to which I ad an increasing pixel count. I then ad another column, representing the time for each pixel measurement. The value of the first cell is the start time of the image in seconds (I usually take the number of seconds after a whole minute, e.g. if the image started at 19:43:32.25 UT the value in this cell is "32.25". If I have a total number of pixels of say 430 (with 430 corresponding pixel brightness values), and an exposure time of 10.05 seconds, then I type this in the cell below it: "=[cell above it]+(exposure/number of pixels - 1)". In our example: "[cell above it] +(10.05/429)".
Then drag this down to the end of the column: the last value now should correspond to the end time of the exposure (in our example, it should be "42.30", i.e. 32.25 + 10.05).

If the raw data graph shows a lot of scatter, it can be useful to apply a running average to the data.

(Note: this approach assumes that the angular motion of the tumbling satellite or rocket stage was fixed over the exposure time in question. In reality, this is not the case. But for short time spans of a few seconds, this can usually be ignored, certainly if the image was taken near culmination of the object. It does introduce some deviation in the result though. Compensating for this makes the exercise a hell of a lot more complicated).

(d) read into PAST and analyse
I then copy the columns with the times and pixel brightness values, and paste them into PAST v.3 (very neat and free statistical software developed by paleontologists. I like it because it is versatile and able to create publication quality vector-format diagrams - the latter ability is something often lacking in such packages).
Press "shift" and select the two columns. Next, under "model" chose "sum-of-sinusoids". Next, a pop-up screen with a diagram appears.
Select "points" under "graph style". I leave "Phase" on "free". You then check the checkbox "fit periods" and click the "compute" button. It will fit a period.
However, I have noted that for some odd reason, the fitted period is not always the best fitting period! Check this by unchecking the "fit period" box, and in the box with the period result, varying the value from the initial fit slightly, after which you press the button "compute" again (leave the "fit period" box unchecked). Look at the R^2 values, and by trial and error find the best R^2 value. This is your actual period.
If your graph shows clearly skewed rather than sinusoidal peaks, than there is a second period interacting with the main period (for example, complex spin motion over two axis, or weaker secondary peaks present). You can try to model this by chosing "2" under "partials".

If you want a nice publishable diagram, press "graph settings" after you are done and adjust the diagram to your liking. Save it as .svg if you want to edit it further in for example Illustrator (as I do), otherwise use one of the other image formats available.

Labels: , , ,

[UPDATED] MUOS 4 has been moved to its operational position at 75 E

MUOS 4 imaged while still in its check-out position at 172 W on 27 September 2015

Since September 2015, I have been periodically covering MUOS 4 (2015-044A), a newly launched military communications satellite in geostationary orbit. It was launched on 2 September 2015 and initially placed at 172 W, just west of Hawaii. This position was temporary: it is the "check-out position" where the satellite is initially placed, well situated with regard to key monitoring stations, to check if it is working okay. It stays there for a few months until this check-out is complete: then it is moved to its operational position. In the case of MUOS 4, it is known (see my earlier post and this unclassified publication) that the operational position assigned to MUOS 4 is at 75 E, south of India.

At the start of December 2015, it was announced that check-out had been completed and that the satellite would be moved to its operational position in the spring of 2016.

When I checked upon MUOS 4 on 4 December 2015 using the 0.51-m telescope of Warrumbungle obs in Australia, it was still at 172 W. For various reasons I did not get to check that position again until a few days ago. When I imaged that location on March 1, 2016, using the same Warrumbungle telescope, MUOS 4 was no longer there. It had been moved somewhere between December 4 and March 1.

This is confirmed by observations of the Russian ISON network. I received two of their MUOS 4 elsets for mid February 2016, which show the satellite drifting westwards at a rate of about 3 degrees/day. From the drift rate I reconstruct from these elsets, I find that the move from 172 W to 75 E started near 23.0 January 2016. At this drift rate, it should have reached its designated operational slot at 75 E 37 days later, on 29.0 February 2016.

[UPDATE 11 March 2016:  On March 8, Greg Roberts in South Africa recovered MUOS 4 at 74.8 E, very close to the expected position]

Labels: ,

Monday, February 29, 2016

Imaging North Korea's new Kwangmyŏngsŏng-4 satellite, and the flash period of its UNHA-3 rb

Kwangmyŏngsŏng-4 on 28 Feb 2016
(click image to enlarge)

North Korea's recently launched new satellite (see a  previous post), Kwangmyŏngsŏng-4 (KMS-4: 2016-009A), is finally starting to make visible evening passes here at Leiden.

Yesterday evening, 28 Feb 2016 near 19:45 UT (20:45 local time), I shot the image above, one of two images showing the satellite passing near the Celestial pole. It is a short exposure of 2 seconds with the 2.8/180 mm Zeiss Sonnar lens on my Canon EOS 60D.

Below is the same image, but in black-and-white negative, showing the trail a bit better:

Kwangmyŏngsŏng-4 on 28 Feb 2016
(click image to enlarge)

The object is very faint (probably near mag +7). It needs a rather big lens (the Zeiss 2.8/180 mm has a lens diameter of 6.4 cm), which unfortunately also means a small FOV. Over the two images, a total imaging arc of ~6 seconds, it however appeared to be stable in brightness with no sign of a periodicity due to tumble. So either it is not tumbling, or if it is tumbling at all it must be a very slow tumble.

Some 16 minutes earlier, near 19:28 UT, I also imaged the upper stage of the Kwangmyŏngsŏng/UNHA-3 rocket (2016-009B) that was used to launch the satellite. This object is brighter and shows a nice tumble resulting in periodic flashes. Below are crops from three images spanning 19:28:32 - 19:28:44 UT. The brightness variation is well visible (the bright star it passes in the first image is beta Umi):

brightness variation of UNHA-3 r/b 2016-009B on 28 Feb 2016
(click image to enlarge)

A fit to the measured brightness variation over these three images shows several specular peaks at regular intervals, with a slightly asymetric profile:

click diagram to enlarge

The fit shown in red is the result of two combined sinusoids: a major period of 2.39 seconds with a minor period of 1.195 seconds superimposed (resulting in the slight asymmetry). Pixel brightness over the trails was measured with IRIS. The data were fitted using PAST.

UPDATE 1 March 2016:

I imaged both the UNHA-3 r/b and Kwangmyŏngsŏng-4 again in the evening of 29 Feb 2016. The sky conditions wer less good, and the pass was much lower in the sky. I used the 1.4/85 mm SamYang lens this time, to get a larger FOV in order to try to capture a larger arc.

KMS-4 was captured on four images (2 second exposures) between 19:19:17 - 19:19:34 UT. It was barely visible on the images, but again the brightness appeared to be stable over this 17 second time span.

The UNHA-3 r/b was also captured, and 3 images (5 second exposures) between 18:58:42 - 18:59:07 UT again showed a very nice flash pattern, fitting (like the observations of Feb 28) a flash period of 2.39 seconds:

click diagram to enlarge

The image below is a stack of these three images. The rocket stage moves from upper right to lower left in the image.

Labels: , , , , , ,

Wednesday, February 24, 2016

A consolidated answer to "Masami Kuramoto" about Resurs P1

The MH17 discussion has become extremely messy. It is a highly politicized topic, with active disinformation campaigns and trolling from both the Russian and Ukrainian sides. It is sometimes hard to discern who is honestly seeking the truth, who is honestly seeking the truth but very naive with regard to sources, and who is actively involved in peddling propaganda and disinformation.

Ever since my expert participation in the Dutch Parliament committee hearing of Jan 22, and ever since I expressed caution on this blog about certain satellite images and factually clearly incorrect statements on satellite positions released by the Russian dept. of Defense, some of the trolling has been directed at me.

I have learned not to bother too much with trolls, but when they actively and tenaciously disseminate disinformation and seriously flawed counter-arguments around one of my analysis, I reserve myself the right to a rebuttal, just to set the record straight.

In a previous post I showed that a satellite image released by Russia, purported to be a Resurs P1 image from 17 July 2014 which claimed to show Ukrainian BUK's in a field near Zaroshchens’ke, is problematic. The viewing angles of the "BUK's" in the image do not appear to fit the satellite-to-location geometry, which only allows clearly oblique viewing angles between 45 and 57.5 degrees with the horizontal, from directions ranging from northwest via west to southwest (for more details, see my earlier post in question). I therefore urged caution with regard to these images.

My post has drawn fire on twitter, notably from a twitter user nicked "Masami Kuramoto". While the nick sounds Japanese, and the twitter account claims to be located in Germany, I have a strong suspicion that the entity behind it is Russian.

Kuramoto's chosen line of attack is by questioning the accuracy of the Resurs P1 orbital information which I used. That orbital information came straight from JSpOC (formerly known as "NORAD"), argueably this world's most reliable source of orbital elements. Kuramoto basically tries to advance a claim that the JSpOC tle's for Resurs P1 are either highly inaccurate or even deliberately doctored, and that the satellite in reality passed along a somewhat different trajectory (but with a similar pass time, to match the time listed in the images), thus advocating for the existence of an imaginary trajectory that allows to reconcile the imaging angles with the published images. In order for this to be possible, it is necessary to argue that the real orbit amounts to a significantly shifted orbital plane, i.e. a shifted RAAN value, compared to the JSpOC published orbit.

In an attempt to argue this position, Kuramoto suggestively quoted from the Space-Track terms of use, taking text out of context to insinuate that JSpOC tle's were not accurate for the task:

Kuramoto kept insisting on his perceived "unreliability"of JSpOC tle's, even after I had set him straight on this:

Let me first elaborate on what I pointed out in the tweets above. Kuramoto tried to capitalize on this warning in the Space-Track User agreement:

This statement is relevant to close encounters (with the risk of collision) of two objects in space. What this statement simply means is that it is unwise to base decisions on debris avoidance manoeuvres solely on published tle's. In such cases, very small uncertainties matter. If one uses tle's produced for the epoch of today to make a prediction on a future position of two objects (say: 3 days from now), that prediction for a moment days from now will have a small uncertainty. SGP 4 after all is only a model. These uncertainties are negligible for other purposes, but for close encounter mitigation they matter. Satellites in Low Earth Orbit move some 7 km/s, so a 0.1 second uncertainty in the time of passing a particular point in orbit, hardly something to bother about under normal circumstances, amounts to a positional uncertainty along the orbit of 700 meter. This might not seem much and for other purposes 0.1 seconds and 700 meter is negligible, but for collision avoidance it matters: it might be the difference between a miss or a hit, certainly because the other object introduces a similar uncertainty (i.e., if both objects have 0.1 second uncertainty, the uncertainty in relative distance is 2 x 700 meter = 1.4 km. So if your analysis says they will safely pass 1.4 km apart, they might in reality collide instead. Or conversely, if your analysis says they will collide, the reality might be that they pass each other at a km distance rather than colliding).

In other words: the warning by JSpOC is only relevant to a very specific situation, and concerns uncertainties that are completely negligible for the subject at hand: the position of a satellite with respect to the viewing geometry of a location on earth. The more so because the latter assessment actually uses a tle with epoch very close to the the time of interest, unlike a collision avoidance assessment of a moment more removed in future. The uncertainty pointed out, in no way can change the viewing angles to the extend that it would solve the discrepancies I pointed out in my earlier post.

Now, this could have been a simple misunderstanding, based on a lack of knowledge and insight in the matter on Kuramoto's side.

Kuramoto however next took it to a new level and suggested that JSpOC might have deliberately altered the orbital elements for Resurs P1 post-fact:
The point is: if JSpOC would have done that, the simple reality is that many people working with these data would notice it. Satellites suddenly would be at different positions than where the JSpOC orbital data would put them. Our tracking network for example, frequently catches Russian satellites as byproduct of our tracking of classified objects. On these occasions we would suddenly note large positional errors in that case, and we would even start to see UNIDS (unidentified satellites, which always have our immediate attention) that next turn out to be Russian satellites in orbits not matching their JSpOC orbit. No way that would go unnoticed.

As for the suggestion that the elements were only retrospectively altered, Kuramoto was a bit shocked to learn next that several of us (including me) actually regularly archive the full JSpOC database of orbital elements. I do so several times each month (for July 2014, I for example have archived elements from July 14 and can compare these with elements for that date retrieved from the JSpOC archive today: they are the same, they have not been restrospectively altered). In a retrospective analysis, altering the elements starting at some given date (or only altering them around a given date) would show up as a sudden change in the elements as well.So no: such a plot is simply not realistic.

After this, Kuramoto nevertheless still wished to cast doubt on the JSpOC tle's:

Notwithstanding my earlier rebuttal, he at first simply restated his already rebutted argument:

That would not do of course, and Kuramoto seems to have realized that. In order to maintain his position, Kuramoto had to grasp the next straw. He next brought up a paper by Kelso et al.:

This paper discusses what factors might introduce predictions that do deviate considerably from reality (with the focus again on the accuracy of data needed for orbital debris avoidance manoeuvres). One such case is when for example a position is based on a tle issued 4 hours ago, but the satellite in question meanwhile has actively manoeuvered to a new orbit. In that case, the predicted position indeed would be incorrect. Kuramoto (of course) tries to seize on that, but in doing so again shows a lack of insight in the matter. Whether a satellite (Resurs P1 in this case) had just manoeuvered can easily be checked: by looking at a series of tle's issued around the time of interest (17 July 2014, 8:32 UT in this case), a manoeuvre around the time of interest would be visible by a sudden change in elements.

For Resurs P1 around 17 july 2014, I did this check (Kuramoto obviously didn't). There is no such change, i.e. the satellite did not make a significant manoeuvre. This can be seen in the diagrams below which depict the evolution of the orbit (from JSPOC data over July 2014). A manoeuvre would show up as a clear discontinuity (a clear sudden change) in either perigee and/or apogee altitude, argument of perigee, inclination, Mean Motion and /or RAAN (and notably in RAAN for Kuramoto's argument to hold). Tampering with the orbital elements around 17 July by JSpOC would show up similarly, by the way. But none of this happens, as you can see below. So again, Kuramoto's next grasp at a straw, is futile, and by now his attempts to argue my analysis away are bordering the pathetic.

post-edit  24 Feb 2014, 15:05 UT:

Kuramoto is still trying to advance his ill-fated argument:

Again, his argument is largely irrelevant. The kind of deviations pointed out are very minor: a maximum error of 9.3 km in position at a given time really will not significantly change the viewing angles. That would need cross-track errors an order of a magnitude larger.

post-edit 24 Feb 2014, 15:40 UT:

Well now, somebody has seen the light it appears:

Labels: , , , ,