Go Back   IceInSpace > General Astronomy > Observational and Visual Astronomy > Observation Reports

Thread Tools Rate Thread
Old 07-10-2013, 11:02 AM
Weltevreden SA's Avatar
Weltevreden SA (Dana)
Dana in SA

Weltevreden SA is offline
Join Date: Apr 2012
Location: Nieu Bethesda, Karoo, South Africa
Posts: 216
Eyewitness account of the SpaceX rocket failure Sunday the 29th

On Sunday Sept 29 I happened to observe the SpaceX Falcon 9 rocket's second-stage LOX fuel venting as the booster stage was midway through its launch of multiple satellites in two sequences. The first sequence successfully placed a number of low-orbit mini- or cubesats. The second launch sequence was supposed to lift a larger satellite into a geosynchronous parking orbit for later repositioning into final orbit. I observed the excess fuel venting with seconds of its commencement. It made a spectacular show, but this was modest compared with what happened to the plume on its second orbital passage 100 mins later.

The Falcon 9 second-stage fuel dump above southwestern Indian Ocean off South Africa commenced 09-29-2013 at ~17:05 UT. Also reported here is a subsequent 2nd orbital passage starting ~18:40 UT. Both events were observed from Weltevreden Farm, S Africa, 33 18’ So, 26h 39m 16s E.

It's been 11 days since the SpaceX event happened. Weltevreden is a remote farm with no internet and no cellphone access. It is as radio and UHF dark as it is visually dark. I came into Internet range only today, hence this late posting:

The two distinct sightings were separated by ~1 hour 40 mins. The first event was also recorded with a cellphone video shot in Mauritius and posted on YouTube. As for the second orbital passage 100 minutes later, I've seen no reports of it at all, even though it was much more spectacular. I call the second passage 'The Plume'.

The Flare

Shortly after sunset on 29 Sept 2013 I was setting up my equipment for a night of visual observing. The charts were out but no scopes were yet on their mounts. The site is very dark with no human-originated lights visible anywhere except for a faint light dome above the nearest village, Nieu Bethesda, 13 km distant in a valley on the other side of a range of hills 8 km E. Seeing was 6/10 with moderate upper altitude turbulence. Transparency 8/10, typical for the season and region. No clouds appeared in the sky till after 03:00 the next morning. Astro dark was achieved half an hour after this observation.

About 17:02 UT a bright light in the sky appeared in the corner of my eye. I glanced up to see a very bright, diffuse, evenly luminous circle roughly 1/3 the dia. of the moon. It was about mag -1 to -1.5 overall surface brightness as compared with nearby Fomalhaut. The visual magnitudes quoted here are estimates of aggregate surface brightness, not point-source. Based on the Flare's behaviour over the next few minutes, I first observed it about 5 to 10 seconds after the event began. Having no idea what this was at the time (and like most astronomers, harbouring dreams of eyewitnessing a supernova at ignition) I sat down at my chart table with the Torres A charts, a notebook, and a pair of 15x70 binoculars.

The Flare began in the the vicinity of the Blanco 1 open cluster in Sculptor. The glow faded out of sight about 20 mins later near 19 Pisceum. The Flare's visual magnitudes were estimated using Fomalhaut, B Ceti (Dubhe), and Alpha Sculptoris. Achenar was still too low SE to be a reliable comparison object.

The Flare appeared as a spherical shell and expanded rapidly. Initially it seemed a soft blueish/yellow hue resembling urban light pollution from mixed sodium and mercury vapour lamps. The Flare was faintly yellow at first, but within another 30 seconds developed a distinct outer halo more bluish than its yellow inner side. After about 3 mins the shell developed a darker core which also expanded, but remained a consistent ~1/4 overall diameter as the cloud expanded. The core-shell interface was a slender yellowish ring approx 1/6 the diameter of the bluish outer corona. The entire cloud was expanding noticeably. The core region also had a bright stellar-like centre approx. mag -1.0 to 0. This 'star' was brighter than Fomalhaut but somewhat diffused by the gas shell surrounding it; it resembled a bright star behind thin cirrus cloud.

Switching back and forth from naked eye to using binoculars, the bright central 'star' appeared more point-like in binoculars. This feature retained the same visual brightness of ~mag 0 to +1 over several minutes. There was no scintillation, although nearby Fomalhaut was scintillating with a noticeable twinkle.

By two minutes into the sighting, the Flare passed ~2 E of Kappa 1 and Kappa 2 Sculptoris, which were still faint in the advancing twilight but detectable just enough to plot the Flare's lateral motion. Around 8 minutes into the event the shell and the darker core within had expanded to their near-final size of ~2.5 dia. Within the first minute the Flare had reached approx the dia. of the moon and increased overall luminosity. By 12 mins into the observation the outer edge had an apparent diameter ~3 times the moon, and its shell-like structure was clearly evident. By 15 mins the soft outer of the ring had muted to a bright gray, having faded to mag +2 to +2.5. The core region remained ~mag 3.5 across 30 arc min, with some hints of blue on the outside and a yellowish inner side. The core and shell both darkened at similar rates as both faded to extinction, although the core never became as dark as the adjacent sky background until finally all of them vanished.

Working backwards from my notes, the spherical expansion rate would have been approx 8.25-8.5 arcmin/min in the first few minutes, slowing to approx 3 arcmin/min in the final five minutes before becoming invisible in the sky glow. The overall structure faded in what appeared to be a linear luminosity falloff affecting both diameter and surface brightness. I calculated the object's vector motion at ~3.6 per min proceeding N from its apparent origin near Blanco 1 (RA -30 dec 00h05m), along the 00h meridian towards the Pisces 'fish' below Pegasus. There is margin for error in the path and vanishing point because of low sky brightness during mid twilight and few usably bright comparison stars. As reconstructed in Chart I attached, the Flare was in near-polar earth orbit.

The final few mins were uneventful compared with the first 15 mins. The Flare had reached a diameter of 2.5x to 3x the moon's diameter and become uniformly faint and colourless, with only a trace of the inner dark core visible. The central stellar-like object disappeared about 15 mins into the 20-min session, though it remained faintly in binocular view. Overall surface brightness descended below ~mag 4. The Flare disappeared from view at an estimated overall brightness of 4.2 to 4.6 in a local sky brightness of just about the same.

The Plume

I went on about my planned observing session. I had calculated the Flare's orbital return time to be ~100 minutes based on the motion rates described above. At 20:30 local (18:30 UT) I moved the table, charts, and binoculars to a spot with unobstructed views from horizon to horizon. By now the sky conditions were limiting visual magnitude vm7.4 to vm7.6 and seeing was 7/10 . The late September Milky Way is roughly 75 elevation at the Sagittarius bulge region. The Milky Way's bar/spiral arm structure is easily traced across the sky, and gives a vivid 3D impression of the bar structure extending from Norma across the bulge to the Scutum star cloud, with the Scutum/Centaurus spiral arm sweeping from the bar tip in Scutum underneath the bulge toward Centaurus; this arm is clearly interior to the Orion Spur where the Sun is located. The remote Perseus Arm on the other side of the bulge originates at the tip of the bar in Norma but doesn't become identifiable as a spiral arm till it emerges from behind the bulge/inner arm region starting in Aquila-Cygnus.

At 18:35 UT a bright streak appeared about 8 above the southern horizon. It entered Musca (~70 S latitude), obscuring the GCs NGC 4372 and 4833. Over the next 20 mins the Plume swept past B Arae, then bisected the rectangle of Telescopium. It proceeded onward a few degrees leftward of the globular NGC 6723 CorAust, thence northeast toward to the Barnard's Galaxy region. By the time the Plume's leading edge reached the Sagg/Capricorn region, the tail end was finally coming into view in Musca. This translates to a total visible sky span of between 120 to 130, rising to roughly 55 altitude above the horizon at the Plume's highest elevation in Sagg.

Seeing a single bright object span Musca to Bernard's Galaxy is the experiences of a lifetime. If it had been a comet, with its overall surface brightness of vm2 evenly spread across the entire length, this phenomenon would arguably rank as one of the great comets—it was only a little less bright than the contemporary accounts of the Great Comet of 1861.

The entire Plume proceeded as a unit along its path, roughly 10 leftward (E) and parallel to the MW disk . This puts it 30 W of a polar orbit passing through Octans. While 100 minutes earlier the path of the Flare approximately followed the 00h meridian, now the Streak's path crossed the 18h meridian at an approx angle of -30 CCW at latitude -57 S, adjacent to the GC NGC 6584.

SpaceX explains the Flare as an expanding outgas cloud after a LOX propellant venting. I wonder about certain details of this explanation. A rocket fuel vent is a nozzle with a directional orientation, which should have resulted not in the spherical shape I saw but rather a cometary shape moving outward along the nozzle axis. A venting is consistent with the subsequent Plume, but not with why the original Flare was so clearly spherical. Given the dynamics of an orbiting gas cloud expanding into space, how did the spherical shape I saw, and which the Mauritius video also shows clearly, become the elongated, pencil-like Plume?

The visual appearance of the Plume was complex. The forward end began as a soft-edged band approx 1.5 wide, which then got thinner from the proceeding tip towards the centre point. There it slimmed into a thin, pencil-like section less than the moon dia (16') across the central 30% of the object's total length. The Plume then widened again into the trailing segment. The trailing tip of the Plume resembled the leading tip in reverse. The central pencil-like central segment was notably sharp-edged and evenly luminous, but starting about 30% outward from the centre, the diffuse leading and trailing segments showed flocculent effects for approx 25% of their length before they diffused into uniform surface brightness in their outer 10% zones. In binoculars, the flocculent regions resembled airline contrails as they deform into patchy knobs before evaporating. The 'contrail' mage is not exact, however, in that the flocculence more resembled gradually undulant cusps frozen in place. I could detect four wave-like cusps on either side of the centre. These waveforms were not sinusoidal: they appeared only on the W facing side of the Plume. The front and trailing tips of the Plume looked to be semicircular, with a rapid falloff to sky brightness at their outermost edges.

There was no repeat passage 100 minutes later. Given the above brightness, size, and path, a third appearance would have been visible over the Western Cape in South Africa/Namibia, though below the horizon for me.

It may not have been supernova-class, but SpaceX provided South and East African skies with a truly memorable event.

=Dana in SA
Attached Files
File Type: pdf Path of Flare 17.10 UT 09-29-13.pdf (52.6 KB, 25 views)
File Type: pdf Path of Streak 18.40 UT 09-29-13.pdf (232.2 KB, 16 views)

Last edited by Weltevreden SA; 11-10-2013 at 09:42 AM. Reason: Added images of paths
Reply With Quote


Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

All times are GMT +10. The time is now 01:35 AM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.