Friday 31 January 2020

California 30 January 12:30 UT: the "space debris" reentry that wasn't

On 30 January 2020 near 12:30 UT (10:30 pm PST), a bright, slow, spectacularly fragmenting fireball swooped over southern California. It was seen and reported by many in the San Diego-Los Angeles area. The video above was obtained by a dedicated fireball all-sky camera operated by Bob Lunsford. The fireball duration approached 20 seconds.

In the hours after the fireball, the American Meteor Society (AMS) initially suggested that this was a Space Debris reentry, i.e. the reentry of something artificial from earth orbit.

But it wasn't.

Immediately upon seeing the video, I had my doubts. Upon a further look at the video, those doubt grew. To me, the evidence pointed to a meteoritic fireball, a slow fragmenting fireball caused by a small chunk of asteroid entering our atmosphere.

A discussion ensued on Twitter, until NASA's Bill Cooke settled the issue with multistation camera triangulation data, which showed that this was an object from an Apollo/Jupiter Family comet type heliocentric orbit with a speed of 15.5 km/s. In other words: my doubts were legitimite. This was not a space debris reentry but indeed a chunk of asteroid or comet.

I've already set out my argumentation about my doubts on Twitter yesterday, but will reitterate them again below for the benefit of the readers of this blog.

My doubts started because while watching the video I felt that the fireball, while slow and of exceptionally long duration, was still a tad too fast in angular velocity in the sky, and too short in duration, for this to be space debris. In the video, it can be seen to move over a considerable part of the sky in just seconds time.

The image below shows two stills from the video 6 seconds apart in time. The fireball passes two stars, alpha Ceti and beta Orionis, that are 35 degrees apart in the sky, and it takes the fireball a time span of about 6 seconds to do this, yielding an apparent angular velocity in the sky of about 5-6 degrees per second. That is an angular velocity that is a factor two too fast for reentering space debris at this sky elevation, as I will show below.

stills from the fireball video, 6 seconds apart, with two stars indicated

Orbital speed of a satellite is determined by orbital altitude. Reentering space debris, at less than 100 km altitude, has a very well defined entry speed of 7.9 km/s. This gives a maximum angular speed in the sky of about 5 degrees/second would it pass right above you in the zenith (and only then): but gives a (much) slower speed (2-3 degrees/second) when the reentry is visible lower in the sky, such as in the fireball video.

To gain some insight in the angular velocity a reentering piece of space debris would have at the elevation of the California fireball, I created an artificial 70 x 110 km reentry orbit over southern California that would pass the same two stars as seen from San Diego.

The map below shows that simulated track, with the object (marked by the green rectangular box) at 70 km altitude and positioned 6 seconds after passing alpha Ceti (marked by the green circle):

Simulated reentry track. click to enlarge
The angular velocity in the sky for a reentering object at this sky elevation suggested by this simulation is barely half that of the fireball. During the 6 seconds it took the fireball to move over 35 degrees of sky passing alpha Ceti and beta Orionis, the simulated reentering object would have moved over only 15 degrees, i.e with an angular velocity of 2.5 degrees/second rather than the 5-6 degrees/second of the fireball.

So this suggested that the fireball was moving at a speed a factor two too high for space debris. This therefore pointed to a meteoritic fireball, not a space debris reentry.

There were other reasons to doubt a reentry too. There were no matching TIP messages on Space-Track, the web-portal of CSpOC, the US military satellite tracking network. A reentering object as bright as the fireball in the video would have to be a large piece of space debris: this bright is clearly not the "nuts and bolts" category but suggests a large object like a satellite or rocket stage. It is unlikely that CSpOC would have missed a reentry of this size.

To be certain I ran a decay prediction on the full CSpOC catalogue with SatEvo myself: no object popped up that was expected to reenter near this date either, based on fresh orbital elements.

The fragmentation in itself, one of the arguments in the AMS' initial but mistaken conclusion of a "space debris reentry", is not unique to space debris reentries. It is also a common occurence with slow, meteorite dropping asteroidal fireballs, especially when they enter on a grazing trajectory. Take the Peekskill meteorite fall from October 1992 for example:

Likewise, while a 20-second meteor is not everyday, it is not a duration that is impossible for a meteor. Such durations (and even longer ones) have been observed before. Such long durations are especially the case with meteors that enter in a grazing way, under a shallow angle.

At the same time, a 20 seconds duration would be unusually short for a satellite or rocket stage reentry. Such reentries are usually visible for minutes, not a few seconds or a few tens of seconds.

So, to summarize:

1) the angular velocity in the sky appeared to be too large for space debris;
2) the fireball duration would be unusually brief for space debris;
3) and there were no obvious reentry candidates.

On the other hand:

a) the angular velocity would match those of slow ~15 km/s meteors;
b) the 20 second duration, while long, is certainly not impossible for a meteor;
c) the fragmentation observed occurs with slow asteroidal origin meteors as well.

Combining all these arguments,  my conclusion was that this was not a space debris reentry, but an asteroidal origin, slow meteoritic fireball. This was vindicated shortly later by the multistation camera results of Bill Cooke and his group, which yielded an unambiguous speed of 15.5 km/s and as a result a heliocentric orbit, showing that this was not space debris but a slow chunk of asteroid or Jupiter Family comet.

In defense of the American Meteor Society (who do great work on fireballs): it is not easy to characterize objects this slow, certainly not from single camera images and visual eyewitness reports. Given the slow character and profuse fragmentation, it is not that strange that the AMS initially (but incorrectly) thought it concerned a space debris reentry. It does go to show that you have to be extremely careful in drawing conclusions about slow moving fireballs: not every very long duration fragmenting fireball is space debris.

Monday 20 January 2020

Testing a new lens for GEO and HEO (SamYang 2.0/135 mm)

The past week brought some clear skies. It also brougt me a new lens, a SamYang 2.0/135 mm ED UMC.

This lens had been on my wish-list for a while, as a potential replacement for the 1979-vintage Zeiss Jena Sonnar MC 2.8/180 mm I hitherto used for imaging faint Geosynchronous (GEO) and Highly Elliptical Orbit (HEO) objects, objects which are typically in the magnitude +10 to +14 range.

The 2.0/135 mm SamYang lens has gotten raving reviews on photography websites, several of these reviews noting that the optical quality of this lens is superior to that of a Canon 2.0/135L lens. And this while it retails at only half the price of an L-lens (it retails for about 460 to 500 Euro).

While I have the version with the Canon EF fitting, the SamYang lens is also available with fittings for various other camera brands.

Focussing is very smooth and easy with this lens. Unlike a Canon-L lens, the SamYang lens is fully manual (both focus and F-stop), but for astrophotography, manually focussing is mandatory anyway. The general build of the lens is solid. It is made of a combination of metal and plastic.

While not particularly lightweight, the lens is lighter in weight than my 1979-vintage Zeiss (which is all-metal and built like a tank, in true DDR fashion). The SamYang has a somewhat larger aperture (6.75 cm) than the Zeiss (6.42 cm), meaning it can image fainter objects. It also has a notably wider field of view (9 x 7 degrees, while the Zeiss has 7 x 5 degrees).

So for me, this seemed to be the ideal lens for GEO and HEO.

And after two test nights I can confirm: this SamYang lens indeed is spectacularly sharp. The first test images, made on January 15 and 16, have truely impressed me. Even at full F2.0 aperture, it is sharp from the center all the way to the edges and corners of the image.

Here is a comparison of the image center and the upper right corner of an image, at true pixel level. There is hardly any difference in sharpness:

click to enlarge

The images below, taken with the SamYang on a Canon EOS 80D, are crops of larger images, all but one at true pixel level.

The first image is a test image from January 15, a nice clear evening. It shows two objects in HEO: a Russian piece of space debris (a Breeze-M tank), and the classified American SIGINT satellite TRUMPET 1 (1994-026A). Note how sharp the trails are (this is a crop at true pixel level):

Click image to enlarge

The next night, January 16, I imaged several geostationary objects (which at my 51 degree north latitude are low in the sky, generally (well) below 30 degrees elevation). While the sky was reasonably clear, there were lingering aircraft contrails in the sky, locally producing some haze. Geostationary objects showed up well however, better than they generally did in the Zeiss images in the past.

The image below, which is a crop of a larger image, is not true pixel size, but slightly reduced in size to fit several objects in one image. It shows the Orion Nebula, several unclassified commercial GEO-sats, the Russian military comsat KOSMOS 2538 (BLAGOVEST 14L), and the classified Italian military communications satellite SICRAL 1B (2009-020A):

Click image to enlarge

The images below are all crops at true pixel level. The first one shows the US classified SIGINT satellite PAN/NEMESIS I (2009-047A), shadowing the commercial satellite telephony satellite YAHSAT 1B. It also shows a number of other unclassified commercial GEO-sats.

PAN/NEMESIS 1 is an NSA operated satellite that eavesdrops on commercial satellite telephony (see my 2016 article in The Space Review).

Note that this image - just like the next images- was taken at very low elevation, and from a light-polluted town center.

click image to enlarge

The image below shows another US classified SIGINT satellite, Mentor 4 (2009-001A), an ADVANCED ORION satellite. It shadows the commercial satellite telephony satellite THURAYA 2 (more backgrounds on this in my 2016 article in The Space Review). At magnitude +8, it is one of the brightest geosynchronous objects in the sky (note how it is much brighter than THURAYA 2):

click to enlarge

The last image below again is a classified US military SIGINT satellite, MERCURY 2 (1996-026A). While 24 years old it is, together with its even slightly older sibling MERCURY 1 (which I also imaged but is not in this image), probably still operational:

Click image to enlarge

After these two test nights, I am very enthusiastic about the SamYang lens. It is incredibly sharp, also in the corners, easy to focus, goes deep (in terms of faint objects), and overall performs excellent. I also like the wide field of view (compared to the 180 mm Zeiss which I previously used to target GEO). Together with the equally well performing SamYang 1.4/85 mm, it might be the ideal lens for imaging GEO and HEO.

Astrometric data on the targetted satellites from these test images are here and here. The astrometric solutions on the star backgrounds in the images had a standard deviation of about 2".

Added 20 Jan 2020:

This last image (reduced in resolution to fit) was taken this evening (20 January) and shows Trumpet 1 (1994-026A) passing the Pleiades:

Click image to enlarge

Sunday 19 January 2020

Imaging Starlink 2

click to enlarge

A new set of 60 Starlink satellites, Starlink 2 (the third launch), was launched by SpaceX early on January 7th. Over the past 10 days, all passes were in earth shadow for my 51 degree North latitude, but as of this weekend, the satellites start to make low visible passes in evening twilight.

Yesterday evening was one of the first opportunities. The Starlink satellite "train", already dispersing as their orbits are raised, would make a pass low south with a maximum elevation at 28 degrees, where they would enter earth shadow.

Conditions were dynamic, with fields of clouds moving in the sky. Initially, the part of the sky where they should be brightest was obscured by a cloud, so I pointed the camera more west and lower in the sky.

The image below is a stack of 65 images, 5 seconds exposure each with 1 second intervals, taken between 17:52:50 - 17:59:15 UT (representing a 6m 25s period), with a Canon EOS 80D and EF 2.5/50 mm Macro lens set at F2.8, 1000 ISO. There is a band of Starlink objects, diagonally from lower right to upper left crossing behind the tree. These are objects in the 'head' of the Starlink 2 main "train":

click to enlarge

When the sky near the satellite culmination point also cleared of the field of clouds, I repositioned the camera to that point and captured the last part of the main "train" tail.

The first image below is a stack of 10 images, taken between 17:59:30 - 18:00:30 UT, representing a 1-minute period. The objects can be seen entering earth shadow at left.

The second image below is a single shot image (5-second exposure) from that series, showing four Starlink objects.

click image to enlarge

click image to enlarge

Near their culmination point, the Starlink satellites were clear naked-eye objects, with a brightness of approximately mag. +2.5 tot +3.0.

The images were taken from the center of Leiden town in the Netherlands, in a twilight sky that suffers quite some light pollution.