Hubble catches stellar explosions in NGC 6984

Nov 08, 2013
Supernovae are intensely bright objects. They are formed when a star reaches the end of its life with a dramatic explosion, expelling most of its material out into space. The subject of this new Hubble image, spiral galaxy NGC 6984, played host to one of these explosions back in 2012, known as SN 2012im. Now, another star has exploded, forming supernova SN 2013ek -- visible in this image as the prominent, star-like bright object just slightly above and to the right of the galaxy's center. Credit: ESA/NASA

Supernovae are intensely bright objects. They are formed when a star reaches the end of its life with a dramatic explosion, expelling most of its material out into space.

The subject of this new Hubble image, spiral galaxy NGC 6984, played host to one of these explosions back in 2012, known as SN 2012im. Now, another star has exploded, forming supernova SN 2013ek—visible in this image as the prominent, star-like bright object just slightly above and to the right of the galaxy's center.

SN 2012im is known as a Type Ic supernova, while the more recent SN 2013ek is a Type Ib. Both of these types are caused by the core collapse of that have shed—or lost—their outer layers of hydrogen. Type Ic supernovae are thought to have lost more of their outer envelope than Type Ib, including a layer of helium.

The observations that make up this new image were taken on August 19, 2013, and aimed to pinpoint the location of this new explosion more precisely. It is so close to where SN 2012im was spotted that the two events are thought to be linked; the chance of two completely independent supernovae so close together and of the same class exploding within one year of one another is a very unlikely event. It was initially suggested that SN 2013ek may in fact be SN 2012im flaring up again, but further observations support the idea that they are separate supernovae—although they may be closely related in some as-yet-unknown way.

Explore further: Compact galaxy groups reveal details of their close encounters

Related Stories

Spiral beauty graced by fading supernova

Mar 20, 2013

(Phys.org) —About 35 million light-years from Earth, in the constellation of Eridanus (The River), lies the spiral galaxy NGC 1637. Back in 1999 the serene appearance of this galaxy was shattered by the ...

Hubble spotted a supernova in NGC 5806

Sep 02, 2012

(Phys.org)—A new image from the NASA/ESA Hubble Space Telescope shows NGC 5806, a spiral galaxy in the constellation Virgo (the Virgin). It lies around 80 million light years from Earth. Also visible in ...

Key link found in Cosmic Distance Ladder

Jun 20, 2013

(Phys.org) —When observing the bright explosion of a White Dwarf star in our neighbouring galaxy last year, researchers from The Australian National University collected the largest ever data set on what ...

Astronomers discover light echo from supernova

Jun 04, 2013

(Phys.org) —Astronomers have discovered light echoing off material surrounding a recent supernova explosion, SN 2009ig. The dust and gas that are reflecting the light are so close to the eruption center ...

Recommended for you

Kepler proves it can still find planets

20 hours ago

To paraphrase Mark Twain, the report of the Kepler spacecraft's death was greatly exaggerated. Despite a malfunction that ended its primary mission in May 2013, Kepler is still alive and working. The evidence ...

User comments : 21

Adjust slider to filter visible comments by rank

Display comments: newest first

thermodynamics
4.3 / 5 (6) Nov 08, 2013
The concept of one supernova influencing another takes my breath away. What an amazing observation. I want to see what they propose as the causal relationship between the two.
Q-Star
5 / 5 (7) Nov 09, 2013
The concept of one supernova influencing another takes my breath away. What an amazing observation. I want to see what they propose as the causal relationship between the two.


It is a wonderful discovery, serendipity defined. It would help to know the spacial separation.
Fleetfoot
5 / 5 (4) Nov 09, 2013
It would help to know the spacial separation.


http://www.cbat.e...vae.html

__SN__ : __Date___ : _____SN_Position____
2013ek : 2013 07 24 : 20 57 53.90 -51 52 24.5
2012im : 2012 07 25 : 20 57 53.92 -51 52 24.8

CBET 3607:

https://www.fcagl...135.html

"Possible interpretations of the two events (2012im and 2013ek) include two
nearby explosions or non-terminal explosion of the same object. Additional
observations are encouraged to confirm the spatial coincidence of the two
events and to monitor the potentially rapid evolution of this latest event."
Q-Star
5 / 5 (6) Nov 09, 2013
It would help to know the spacial separation.


http://www.cbat.e...vae.html

"Possible interpretations of the two events (2012im and 2013ek) include two
nearby explosions or non-terminal explosion of the same object. Additional
observations are encouraged to confirm the spatial coincidence of the two
events and to monitor the potentially rapid evolution of this latest event."


Yeah, I had already looked up the angular separation, I was wondering if they had any idea of the radial separation so as to know how close the two objects were to each other.
RealScience
5 / 5 (2) Nov 09, 2013
@Q-star- Radial separation would be very hard to determine directly.

The galaxy is 180 M light years away, so we can calculate a minimum spatial separation based on the listed angular separation: 51 52 24.8 - 51 52 24.5 = 0.3 arc-seconds or 0.3 * 180 M * pi/180 / 60*60 = ~250 light years minimum distance apart.

If the 0.3 arc seconds were accurate, astronomers wouldn't be suggesting influencing each other within a year of our detection time. So I'd guess that even for angular separation the error bars are big enough to make the calculation useless.

And I don't see other direct means for measuring. These things are very bright, so just light echoes from gas clouds suggests itself, but they'd need to map these to higher accuracy than the explosions to help - perhaps that will be possible in a few decades.

@Fleetfoot - Did I make any silly mistakes in the above calculation? Do you happen to know the measurement accuracy?
Q-Star
5 / 5 (6) Nov 09, 2013
The galaxy is 180 M light years away, so we can calculate a minimum spatial separation based on the listed angular separation: 51 52 24.8 - 51 52 24.5 = 0.3 arc-seconds or 0.3 * 180 M * pi/180 / 60*60 = ~250 light years minimum distance apart.

If the 0.3 arc seconds were accurate, astronomers wouldn't be suggesting influencing each other within a year of our detection time. So I'd guess that even for angular separation the error bars are big enough to make the calculation useless..


I'm thinking it is more like 0.36 ly angular separation, which is why the radial component is the only one which will give any meaningful indication of their spatial separation. Angular separation they are basically right on top of each other. sin (360*3600 / 0.15) * 180*10^6
RealScience
5 / 5 (2) Nov 09, 2013
@Q-Star - positions in astronomy are generally given in degrees, and the declination of the center of NGC 6984 is -51 degrees 52 minutes 14 seconds so Fleetfoot's figures sure look like a difference of 0.3 arc-seconds.
To check on measurement uncertainty I clicked on the position reference in the link he sent, but it asked me for a password...

I'll dig around a bit...
Fleetfoot
5 / 5 (3) Nov 09, 2013
The galaxy is 180 M light years away, so we can calculate a minimum spatial separation based on the listed angular separation: 51 52 24.8 - 51 52 24.5 = 0.3 arc-seconds or 0.3 * 180 M * pi/180 / 60*60 = ~250 light years minimum distance apart.


That's a good estimate, a spreadsheet gave 262LY (the difference in RA is negligible).

If the 0.3 arc seconds were accurate, astronomers wouldn't be suggesting influencing each other within a year of our detection time.


It would rule out one causing the other but not a third event say several centuries before triggering both.

@Fleetfoot - Did I make any silly mistakes in the above calculation?


No, you got it right (or we both made the same mistake).

Do you happen to know the measurement accuracy?


I don't know the accuracy but since they also consider that it might be two novae from the same star, it seems the uncertainty is higher than the separation.
RealScience
5 / 5 (3) Nov 09, 2013
@Q-Star - the angular resolution limit of the Hubble is 0.05 arc-seconds, or on the order of 50 light years at that distance.
One could do ~5x better with the LBT or ~10x better using both Keck telescopes together, but since I don't think that 2012im was measured that way and is probably not still imageable, we're probably out of luck on direct observation pinning down the separation (and even the two Kecks couldn't resolve better than the distance from our sun to Alpha Centauri).

But there are a lot of very smart people out there so someone may well find a trick, especially if they are close enough be orbiting each other.

@Fleetfoot - thanks for the math check!
Q-Star
5 / 5 (6) Nov 09, 2013
@Q-Star - the angular resolution limit of the Hubble is 0.05 arc-seconds, or on the order of 50 light years at that distance.
One could do ~5x better with the LBT or ~10x better using both Keck telescopes together, but since I don't think that 2012im was measured that way and is probably not still imageable, we're probably out of luck on direct observation pinning down the separation (and even the two Kecks couldn't resolve better than the distance from our sun to Alpha Centauri).

But there are a lot of very smart people out there so someone may well find a trick, especially if they are close enough be orbiting each other.

@Fleetfoot - thanks for the math check!


I used the sin of 1/2 the angle times the long leg to get the separation. The standard parallax small angle formula in reverse.

The angular resolution is not the tricky part of measuring the radial distances. It would be the photometric & spectrograph data of the individual objects used to determine radial distance.
Fleetfoot
5 / 5 (3) Nov 09, 2013
Q-star :
I used the sin of 1/2 the angle times the long leg to get the separation. The standard parallax small angle formula in reverse.


No problem but you posted: "sin (360*3600 / 0.15)" and it should be sin (0.15 / 360*3600 ) if your calculator is working in degrees, maybe just a typo?

RealScience:
the angular resolution limit of the Hubble is 0.05 arc-seconds, or on the order of 50 light years at that distance.


I think the difficulty may be in aligning the images, what reference points would be used?
Q-Star
5 / 5 (7) Nov 09, 2013
No problem but you posted: "sin (360*3600 / 0.15)" and it should be sin (0.15 / 360*3600 ) if your calculator is working in degrees, maybe just a typo?


No problem as long as ya want a correct answer. I'm sitting here and having a good laugh at myself. Ya both were correct.

@Realscience,,,

I'm trying to type and laugh at the same time. Your distance was quite right, about 4 orders of magnitude than me own.
RealScience
5 / 5 (2) Nov 09, 2013
I think the difficulty may be in aligning the images, what reference points would be used?


I was pointing out that even in the perfect case the HST would not be sufficient to get below ~50 light years. The worst case should be roughly the 0.05 arc-seconds resolution for each object, plus the 0.05 to 0.1 arc-seconds of pointing accuracy of the HST on repeat exposures (which can use the same guide stars), for ~ 0.2 arc-seconds = ~~ 200 light years uncertainty in the distance between them.

But we're still not up to 0.3. Another source of error could be that a super nova is bright enough to saturate a few pixels and thus might not show up as as point-like as something dimmer. In that case a repeated observation could shave e a bit off the uncertainty.

There are some bright quasi-points in the galaxy that are within a few arc-seconds, which could help with aligning images, so one might be able to shave more.

But it's still in the ~~100 ly range, which doesn't help.
yyz
5 / 5 (6) Nov 09, 2013
The observing proposal for HST imaging of SN 2013ek notes that the supernova is less than 0.4 arcsec from SN 2012im. In addition, the Observing Description notes "The 0.04" pixel scale of WFC3/UVIS will constrain the position of supernova and host cluster within ~12 pc".

http://www.stsci....3505.pdf
RealScience
5 / 5 (1) Nov 10, 2013
Thanks, yyz!

Summary: The HST proposal has the distance to the host galaxy at 65 Mpc, which is about 18% further than the 180 Mly used above. So 2013 ek was already known to be within 400 light years of 2012im, and proposal's Hubble time will have pinned 2013ek's position down to 40 light years, leaving the main uncertainty as 2012im's position.

Follow-up time has been requested for observations after the glare of 2013ek fades, so we are likely to know more in roughly one year.
cantdrive85
1 / 5 (6) Nov 10, 2013
It was initially suggested that SN 2013ek may in fact be SN 2012im flaring up again, but further observations support the idea that they are separate supernovae—although they may be closely related in some as-yet-unknown way.


If it's the same object I imagine that would cause quite a problem for the "dying star" guess, shedding it's H after already shedding it's H and He would be quite the feat.
If two separate objects, the fact that stars tend to form along BC filaments would provide for a means of a connection between the two events. Being such a notion would support the EU POV, it is likely the standard fools will link it to DM or some spooky action at a distance nonsense.
RealScience
5 / 5 (2) Nov 10, 2013
If two separate objects, the fact that stars tend to form along BC filaments would provide for a means of a connection between the two events. Being such a notion would support the EU POV, it is likely the standard fools will link it to DM or some spooky action at a distance nonsense.


@cantdrive - If two separate objects, the fact that stars form in clusters through the gravitational collapse of clouds would provide a means of connection between the two events - if a star were getting close to the super nova stage, a very close super nova could trigger the star's final instability.

The fact that you would miss an obvious conventional explanation and thus think that something supports your point-of-view should show even you how unreliable your deductions are and thus where the nonsense is most likely to be.

cantdrive85
1 / 5 (6) Nov 10, 2013
If one event triggered the other and we know there was about a year between the two events, that should place some constraints on the distance between the two objects. It seems it would put the stars VERY close to one another, a binary indeed but still extremely close in astronomical terms. I acknowledge the conventional view, but I also must consider the likelihood of such proximity of two stars capable of going SN.
RealScience
5 / 5 (1) Nov 10, 2013
@cantdrive: And I acknowledge that the conventional view is not invariably correct and should not be considered infallible. None-the-less, when I come to a different result than the conventional view, I first check my math, I then check my logic, and I then check assumptions, and if I haven't a mistake, I then check my understanding of the question, and I then check that I understand how the conventional view reaches its result. Only then do I make a comment on the alternate view, and even then I try to acknowledge that it is merely an alternate view and that the conventional view could be correct.
RealScience
5 / 5 (1) Nov 10, 2013
Regarding the possibility of close supernovae simply by chance, clusters can form a lot of big-enough stars at close to the same time. Such stars typically live a few million years, so if a cluster forms a few hundred supernova candidates, even if there is no other correlation the chance of a second super nova from the same cluster detected within one year is on the rough order of 1/10,000.
Now how many supernovae are detected? It is currently well over 100 per year, so even if there is no correlation between the supernovae it is not THAT surprising that we have found one (and so far only one) random pairing that close in space and time.

One clue will be in the statistics - if SNs in the same cluster within 10 years apart are ~10x more common than those 1 year apart, then it is probably a random pairing. However if we find significantly more 1-year pairs than the 10-year pairs would predict, then there is probably either one causing the other or something causing both.
RealScience
not rated yet Nov 10, 2013
If a Supernova releases 10^44 Joules with spherical symmetry, then an olded Wolf-Rayet star 1/2 light year behind it (from our perspective) would receive ~5x10^37 Joules, or very roughly 10^6 seconds (11 days) of its output. If 11 days worth of a bright star's output arrives over many days I wouldn't expect it to affect the star's core.

Ejected matter can reach ~0.25c, so 1/5 ly would give 1 year detection difference, and a month's worth of energy, but it would arrive over the course of months.

However even a 20-degree beam angle GRB from a type 1C that happened to hit a 1 ly away Wolf-Rayet star would deliver 100 times the energy, and deliver it in seconds. 3 years output delivered to one side of an unstable star in a matter of seconds could easily push it over the edge.

Binaries typically orbit equatorially and GRBs are polar beams, so it would likely be a neighbor rather than a sibling star. But would that be more likely that coincidental close bursts?

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.