SN 2014J

OK Chris!

That's a good idea; we'll migrate to the new topic title. See my last entry in “Possible Supernova in M82”

Reply to Gary,

Your point number “2” is very cogent. The thing hasn't been getting lead from the front. That's just the comment I was inviting. On your point “1”...terrible weather here in the UK also. I've just been very luck with several brief windows of opportunity. May not get any more.

Bill.

"... There seems like a slight nebulosity around the SN. I'll bet this is from all the galactic dust in M82, and the billions of stars in it that seem so close from our distance. Has anyone else noticed this?" Yes, I have. It has made me take a bit longer than usual to do visual estimates. So far I've been able to get 5 since the discovery, and now the weather is bad. The relative lack of attention has puzzled me also. SN 2014J is my first priority target now. Also, I haven't seen anyone posting about amateur spectra as w/ NOVA DEL 2013. I enjoyed reading those as a way of learning about the field and the astrophysics of the expanding shell etc. (So far I am strictly a visual observer but hope to do spectroscopy one day). Does anyone know if many/any amateurs are posting spectra for the SN?

Chris Maloney

Lots of images here:

  http://www.rochesterastronomy.org/sn2014/sn2014j.html

To:  Group

Here are my numbers for SN 2014J using a ST-10XME and 33Cm reflector @ F/7.4:

26Jan14:     B=12.2, V=10.9

29Jan14:     B=11.95, V=10.6, R=10.02, I= 9.7

These are not-transformed, but I will transform and resubmit them soon using Lesvephotometry software.

James Foster,                                                                                                                                      Los Angeles, CA

A collection of (amateur) SN2014J spectra can be shows here:

http://www.astrosurf.com/aras/Aras_DataBase/Supernovae/SN2014J.htm

(it is an international contribution)

Note the quality of investigaion with amateurs setup (time evolution, high quality spectrophotometry, ...)

Christian Buil

No one has ever looked at any supernova with long time series of fast photometry, so we do not know whether any supernova actually does or does not display fast variations (i.e., dips or flares).  Well, certainly, *theory* would be surprised to find fast variations because the expanding shell is so large, and large objects 'cannot' change their brightness on a fast time scale.  (So, a shell one light-day in size should not vary faster than one day or so.  There are exceptions to this rule, perhaps most likely in this case if some small bright spot appears somewhere on the surface of the shell before cooling and fading fast.)   This theory argument also applies to nova when in the optically-thick shell phase, so no one expected that novae around their peak in eruption should show fast variations, and so no one looked.  But the first time that anyone looked at a nova with a lot of fast time-series photometry (for the U Sco eruption in 2010), we found hour-long flares.  This surprise to 'theory' still has not been explained.  No one knows whether supernovae can also show similar flares (or maybe dips).  Theory is all fine, but let us see what reality is.

Fast variations in supernovae cannot be too large in amplitude or too frequent, otherwise any of many groups would have previously picked them out (say, as scatter about some light curve template) even with their one-or-twice a night light curves.  So we are looking for moderate or small amplitude (say, <0.25 mag or so) variations.  This forces us to get time series where the individual images have a statistical uncertainty of 0.01 mag or better (S/N ratio of >100 or so).  The not-high-frequency of the phenomena means that we have to accumulate a long time series to have a hope for catching the events.  Professional scopes cannot allocate so much time to such a program, and certainly, no allocation committee would grant time for such an 'impossible' possibility.  This means that AAVSOers are uniquely and perfectly set up to test out whether supernovae have fast variations, because we can devote many whole nights to sitting on one target.  If the flares/dips are uncommon, then we have the responsibility of providing convincing evidence that any claimed event is real.  For example, we can imagine all sorts of artifacts that could be interpreted as a dip or a flare, like time-variable seeing where the amount of light entering the photometry aperture from the surrounding M82 galaxy changes as the seeing changes and might be thought to be a change in the supernova.  For uncommon events, the best way to convince people is to have two completely independent observers simultaneously report the same flare or dip.  This means that we need multiple people following SN2014J at any given time, so any flare or dip can be proven by mutual confirmation.  So again, the many AAVSOers are perfect because we can marshall many observers round the clock.

I don't know what we'll find, but then no one has ever looked in this way at a supernova before.  Finding a fast flare or dip would be big news, partly because it is so startling and a challenge to theory, but also because it might represent an new phenomenon not realized before.  But before we declare a new phenomenon, we have to have strong proof, and that means multiple simultaneous observations with high photometric accuracy for which we can rule out possible artifacts.

On looking over the light curve to date, I see the best data sets (for purposes of lookiing for fast flares or dips) to be the parallel long time series from JD 2456681.58 to JD 2456681.83 (from UJHA and GJP).  This is great, because we can follow along minute-by-minute, looking for simultaneous flares or dips.  In this case, there are none at the 0.015 mag level.  So this is a good result for a Type Ia supernova near peak.

There are the usual variety of issues that need considering.  The most blatant is that individual observers have self-consistent light curves that parallel each other, but with some offset.  For the JD 2456681.58 to JD 2456681.83 case, the offset is close to a quarter of a magnitude.  Other differences are at the half magnitude level for some observers.  I can see no problems with the individual observer's comparison stars, so very-likely the issue is simply that different observer CCD+Filter+Scope combinations give a bit different color sensitivity, so as to produce a significant color term between the SN and its comparison stars.  This is all normal, and it is easy to work around, conceptually just by adding some arbitrary constant to each observer so that all their time series lie on top of each other.  We are only looking for time changes, so adding a constant is irrelevant.  However, an important implication of this is that we need *long* time series.  In an extreme example, for the purposes of looking for fast flares/dips, a single observation a night is useless, because to find a fast event we can only compare that observation to others within a few hours, and then we won't kow whether any difference is due to a flare/dip or simply due to the usual color-term offsets.  [Note, single nightly observations have other good uses.]  So short time series likely will not be of much help for discovering fast flares/dips.  In practice, what I'd recommend, if you are up to it, it to simply put your CCD on autopilot on SN2014J just after it gets dark and keep on it all night (or as long as you can).

Many nights have long times series (mostly in CV), and we see smooth rises, but there are occassional features.  For example, what about the fading by ~0.1 mag from 2456680.64 to 2456680.71, and then there appears to be a short flare centered on JD2456680.75?  Are these real flares?  It is easy to imagine the real and insidious effects of variable seeing changing the M82 galaxy flux inside the photometry aperture for the SN, but not for the comparison stars, hence resulting in a light curve with flares/dips corresponding to random changes in seeing.  About the only way we can convince people would be to get some completely independent and simultaneous measure of the same flare/dip that shows the exact same light curve.  (The idea is that it is rather unlikely for two widely separated sites to undergo the exact same fluctuations in seeing over an extended time.)  In this case of the JD ~2456680.7 candidate flares, we do not have any parallel long time series to test or confirm/deny the flares.  We do have many observations from the same time in a variety of filters (mostly V and B), with these patchy data not showing the features in the CV time series, so I conclude that the evidence for these candidate flares is much too weak.  But this example highlights the importance of having multiple observers simultaneously making long time series.  Just with two or three people (both in Europe and North America) working all night every clear night, we can cover most of the time with simultaneous time series, and then any flare/dip will be obvious and convincing.

Hi Gary,

These features are typical of most SNe, and are due to both the energetics of the explosion (what is causing the flux, basically Ni56) as well as interior absorption (mostly Fe56).  Michael Richmond gave a good description of this during his Supernova Chat session:

http://www.aavso.org/supernova-photometry-chat-dr-michael-richmond

that I highly recommend reading.

Supernova are another one of those categories of difficult objects on which to perform photometry.  They are almost always in regions with background galaxy contributions, and their spectral energy distribution is nothing like a black body, and is constantly changing, so transformation is difficult.  Nature keeps throwing these pathological objects at us!

Arne