SUNSET: 
16An inconspicuous star in the constellation Sagitta, the Arrow, may become one of the most luminous objects in our sky later this century.
Louisiana State University
An inconspicuous star in the constellation Sagitta, the Arrow, may become one of the most luminous objects in our sky later this century. According to Bradley Schaefer (Louisiana State University), the 11th-magnitude variable star V Sagittae will outshine Sirius and maybe even Venus, around the year 2083 — despite its distance of some 7,500 light-years.
“It’s going to be an awesome sight,” Schaefer said at a press conference at a meeting of the American Astronomical Society in Honolulu. The celestial fireworks aren’t expected until 2083 or so, though, so only the younger readers of Sky & Telescope may have a change to witness the event.
V Sagittae is a cataclysmic variable, a system in which a main sequence star is losing mass to its white dwarf companion. Cataclysmic variables pepper the sky, but V Sagittae is special because the main sequence star is so massive: 3.3 times the Sun’s mass, which makes it 3.9 times the mass of the white dwarf. As a result, the star is losing its outer layers at a prodigious rate of some 6 Earth masses per year.
According to Schaefer and his colleagues, the situation is quickly running out of hand.
Old photographic plates at the Harvard College Observatory, combined with archival data from the American Association of Variable Star Observers (AAVSO), reveal that V Sagittae has been gradually brightening, gaining 2.5 magnitudes (a factor of 10 in brightness) since its discovery in 1904. According to the team, which has submitted the results to Monthly Notices of the Royal Astronomical Society, this indicates a ten-fold increase in the mass falling onto the white dwarf.
As the larger star loses mass to the white dwarf, the binary pair’s orbit should tighten. Indeed, a careful comparison of older and more recent observations show that the two companions eclipse each other from Earth’s perspective, and their orbital period is decreasing. Right now, the stars orbit each other every 12.34 hours, but this value is decreasing by about 0.015 seconds per year.
According to Schaefer and his colleagues, the mass loss is currently doubling every 89 years, an exponential increase that will end catastrophically in the second half of this century. Sometime between 2067 and 2099, the main-sequence star will be ripped apart, dumping about one final solar mass worth of gas onto the white dwarf over just a few days.
The result, which Schaefer describes as a “merge burst,” would maintain a luminosity between that of a regular nova and a supernova explosion for weeks on end. “What remains,” he says, “is a red giant star with a hydrogen-burning mantle and a white dwarf core.”
“It’s an interesting suggestion, that may even be true,” comments astrophysicist Gijs Nelemans (Radboud University, the Netherlands). However, he cautions that the team may be putting too much confidence in rather simplistic theories about the future development of the mass transfer. “I don’t believe that their prediction can be as precise as they suggest,” Nelemans says, “but it would of course be great if the current trend will indeed continue.”
Schaefer expects that future observations will lead to more precise predictions of V Sagittae’s showtime. But he is quite sure that “people the world over […] will see a wondrous guest star shining as the brightest in the sky for a month or so.”
Comments
Don-Brundage
January 9, 2020 at 5:11 pm
When I first saw reports of this, I was wondering why no one was discussing a Type 1a supernova, rather than an unusual nova in a few decades. Then I dug further and saw that the white dwarf's mass was less than half of Chandresekar's limit, so V Sagittae has a long way to go before blowing up in that manner.
I am curious how long the red giant phase would continue before ending in a Type 1a supernova. Decades? Millennia? Probably will take some serious computer modeling to answer that one.
You must be logged in to post a comment.
Brad-Schaefer
January 16, 2020 at 9:46 pm
Yes, you have a neat idea that V Sge might have a classic supernova event at some time. Such would be much more exciting than the predicted merge burst. After all, the white dwarf in V Sge might be now around 0.85 M_sun (far below the Chandrasekhar limit of 1.4 M_sun), but towards the end, the companion will end up dumping over 2 solar masses of solar-composition-gas onto the white dwarf, so it is easy to expect that the white dwarf will sooner-or-later go over the Chandrasekhar limit and necessarily become a Type Ia supernova. Ah, I originally thought the same as you. But in the usual checks with world-class theorists, it quickly became apparent that the mass falling off the companion star will mostly not settle onto the white dwarf. The entropy and energy of the falling matter is such that it cannot settle onto the white dwarf. Rather, falling material will be either blown off in the ferocious wind already blowing or settle into an envelope centered on the white dwarf, including a hydrogen burning layer. So the effective mass of the degenerate core will not rise to the Chandrasekhar limit and we get no Type Ia supernovae. Oh well.
You were also asking the reasonable question as for how long the final merged star, settling into a red giant configuration, would last before it becomes a supernova. (Actually, this path is closer to a 'core collapse supernova' than to a Type Ia supernova.) Well, if we are hoping for a supernova again, then we are out of luck again. After the merger, the combined star cannot be more massive than its two combined components of 0.85 M_sun plus 3.3 M_sun, or 4.15 M_sun or so. (Actually, with the massive ongoing winds, much mass will be lost from the system, so the final merged total must be substantially less than 4.15 M_sun. Details of the wind are hard to know to any accuracy as we get close to the merge burst.) I have not done the calculation for how many more years the merged-V-Sge can last, but I expect that it has another perhaps a fraction of a million years till its evolution runs to the end. Now, for the giant star to produce a core collapse supernova, the mass must be greater than something like 8-10 M_sun. So the V Sge merged system will certainly never have enough mass to 'go' supernova.
You must be logged in to post a comment.
Tom Hoffelder
January 10, 2020 at 1:10 pm
No prob, I'll only be 137!
You must be logged in to post a comment.
Al-Schwarz
January 10, 2020 at 4:54 pm
I'll be 126....I marked my calendar!!!
You must be logged in to post a comment.
Andrew James
January 10, 2020 at 9:44 pm
I do view this recent announcement with significant scepticism based on both the timing of the American Astronomical Society meeting, the degree to rely on numerous broad-based assumptions, and the sheer misinterpretation by the media in understanding even what is the actual basic mechanism in novae production. (Among a dozen astrosites I see gross errors and a lack of understanding with the reportage.) Moreover, Professor Bradley Schaefer said in a press release. “Over the next few decades, the star will brighten rapidly. Around the year 2083, its accretion rate will rise catastrophically, spilling mass at incredibly high rates onto the white dwarf, with this material blazing away. In the final days of this death-spiral, all of the mass from the companion star will fall onto the white dwarf, creating a super-massive wind from the merging star, appearing as bright as Sirius, possibly even as bright as Venus.” This is a oversimplification that was likely made to make it more digestible to the public. Sure. Some merge event might be predicted to occur in 2083, but it is far better the possibility that atmosphere enveloping the white dwarf might ignite beforehand. Many different kinds of astrophysical phenomena could also happen between now and 2083 to make the prediction invalid.
By simple definition, novae are binary systems containing a white dwarf (WD) that accretes material from a nearby companion donor star. For main sequence or sub-giant donors, mass-loss occurs via Roche-lobe overflow and is considered a cataclysmic variable (CV). Once a critical quantity of material has accreted onto the white dwarf surface, the conditions for nuclear burning occurs leading to a thermonuclear runaway under the degenerate conditions of its white dwarf.
To quote a recent March 2019 paper (arXiv:1903.05868v1) by Selvelli1 & Gilmozzi : "Classical novae (CNe) are members of the class of cataclysmic variables (CVs), that is, close binary systems in which a white dwarf (WD) accretes hydrogen-rich material through an accretion disk from a low-mass, near-main-sequence companion. The "classical nova" phenomenon is a thermonuclear runaway (TNR) event that occurs when the accreted mass on the surface of the WD is large enough for the pressure at the base of the semi-degenerate shell to initiate nuclear reactions;"
A novae brightening is a consequence of that mass ignition, resulting in an ejected nebula, and destroying the accretion disk. A recurrent nova is formed when the mass transfer repeatably creates a new accretion disk and another thermonuclear event. Novae generally eject about 0.001 to 0.000001 solar masses at velocities between 500 to 5000 km/sec, and last days to several months before disappearing from view.
Note: If you would like to know of the dynamics of this system, then this arXiv 2003 paper (with figures) "A limit cycle model for long-term optical variations of V Sagittae: The second example of accretion wind evolution" by Hachisu & Kato might help. (See arXiv:astro-ph/0308065) Also V Sagittae is not unique , as it forms a particular sub-class of CV variables. They include : V617 Sgr, HD 104994 and WX Cen (WX Cen is a copycat of V Sge.)
Note 2: In 1932, P.M. Ryves in his MNRAS paper "A remarkable variable star (V Sagittae)" even stated: "The possibility the V Sagittae maybe a nova or closely allied to that class…" According to this paper V Sge was discovered by L. Ceraski in 1902 NOT 1904
Note 3: The figure with the arrow showing where V Sge is incorrect. It should be at the right of the nearby star not left, and follows almost exactly the direction of the asterism. Also Sagitta's outline doesn't actually look like that!
You must be logged in to post a comment.
Rich
January 11, 2020 at 2:53 pm
Regarding Note 3 from AJames:
Could Sky & Telescope post a correct diagram of Sagitta with V Sge?
You must be logged in to post a comment.
Rich
January 13, 2020 at 9:58 am
Here's a link to the area in AladinLite:
http://aladin.unistra.fr/AladinLite/?target=20%2020%2059.708%2B21%2004%207.82&fov=29.94&survey=P%2FDSS2%2Fcolor
You must be logged in to post a comment.
goodricke1
January 11, 2020 at 3:28 pm
Excellent summary AJames, many thanks.
You must be logged in to post a comment.
Monica Young
January 13, 2020 at 11:46 am
Regarding Note 3...the chart provided by Louisiana State University isn't very detailed, so it's difficult to say if the arrow isn't pointing at the right star. In any case, we'll post a Sky & Telescope chart tomorrow (Jan. 14).
You must be logged in to post a comment.
Andrew James
January 13, 2020 at 9:36 pm
Umm… Alpha to Gamma Sagittae (Top star of the tail of the arrow's asterism and the point of the arrow.) are 4.7 deg in PA 71 deg apart. V Sge lies a further 5.2 degrees in PA 71 deg from Gamma Sge - a straight line by any definition. V Sge is at RA 20h 20m 14.6s +21deg 06' 08" I did measured the chart above, and V Sge is 53px right of the arrow, and as this little figure is 250px in total width, the error is visually obvious.
IMO, the prediction of this event in the 2080s is to the novice seemingly like peering into a crystal ball than rational science. If we want to promote astronomy then every effort must focus on the verifiable or undisputed facts, if only just to dull the uncertainty of the bold prediction. All this little LSU figure does is to leave confusion, especially in the confusion that the LSU source provides in their own press release (https://www.lsu.edu/physics/news/v_sge-press_release.php) e.g. Schaefer said at the AAS meeting; "In the final days of this death-spiral, all of the mass from the companion star will fall onto the white dwarf…" but the press release also says: "Inevitably, this in-spiral will climax with the majority of the gas in the normal star falling onto the white dwarf, all within the final weeks and days." There is a significant difference between saying "all" or the "majority" of gas isn't there? (Reading the technical details pdf in the LSU link some of the commentary has been misinterpreted by the media.)
According to their ppt presentation the whole scenario about the 2083 date relies on "Literature and all past experience: Not applicable."! Gulp!! I think we now need to see the whole peer review paper before taking this news as gospel.
Note 1: The orbital period was discovered in 1904 not the variable star, which was in 1902.
Note 2: Days of thinking about this reminds me of the ol' Binary Star Model of PNe Formation from the 1980s, where a red giant envelope engulfs its white dwarf secondary companion, creating a nova event, then followed by forming a planetary nebula with a central white dwarf binary system illuminating the gas by their UV radiation.
You must be logged in to post a comment.
Monica Young
January 15, 2020 at 11:30 am
We have created a diagram showing the location of the star (though it's still faint, of course), as well as a black-and-white finder chart that can be downloaded as a PDF.
You must be logged in to post a comment.
Brad-Schaefer
January 15, 2020 at 11:50 pm
You are rather quick to condemn the science based on inaccuracies appearing only in various press sources. (But Schilling's note above is fine.) For example, you make a big deal out of your trivial and meaningless confusion between 1902 and 1904, as if it affected the science. (Actually, Ceraski's first observation was in 1902, his realization of variability and the publication was in 1903, and the first usable magnitude was from 1904.) And you talk long about nova events, apparently as if such an event will intervene before 2083, and are representing this as a "gross error". Actually, it really would not change anything significantly if V Sge does have nova events, but we know for sure that it cannot, because it is far above the accretion regime where the hydrogen is steadily burning and even puffing up into a shell. Well, neither the press release nor the S&T article above call the merge event as a nova, although aspects of the V Sge merge burst can be comparable to ordinary novae (like the appearance to naked eye observers and to the energetics). So all is fine.
You point out that there is a (very small) group of CVs called the "V Sge stars", so V Sge suddenly does not sound so unique. Ah, but all the other "V Sge stars" have their mass ratio less than or equal to 0.5 (Oliveira & Steiner 2007, A&A, 472, L21), while V Sge is alone with its mass ratio of 3.9. This difference is a big deal, because all known CVs have a mass ratio less than 1.0 or so, while the dividing line for runaway accretion instability is for a mass ratio of 1.0. So V Sge stands starkly alone amongst all known CVs as having a mass ratio much greater than unity, and this dominates the entire evolution, so V Sge has a unique evolution. (And it is exactly this uniqueness as to why such a case has never been seriously handled previously in the literature, which is why we cannot simply get an evolutionary path by looking in old papers.) Perhaps this is just saying that you were fooled by a poorly-named 'class' of CVs.
So in all, you are not showing any "gross errors".
You must be logged in to post a comment.
Andrew James
January 16, 2020 at 7:39 pm
Thanks for this response. I think I'm not condemning the science behind this story but in how that science has been generally reported. An example "gross error" is like this comment on SciFi Wire (https://www.syfy.com/syfywire/-a-star-could-explode-in-2083), where it says: "You can’t see the star V Sagittae (V Sge) yet. In fact, you’ll have to wait quite a while to see it - when it explodes in 2083. The last time something like this happened was a hundred years ago, and no one who was around then could image that one. Now a team of scientists have figured out around when V Sge will go total supernova (unlike Betelgeuse)." Why does the confusion exist? To quote the AAS Press Release it says "The "explosion" will be somewhere between a nova and a supernova." I also see such failure more as not stating or defining what a novae is or the processes that cause them.
Another is Universe Today reportage (https://www.universetoday.com/144529/forget-betelgeuse-the-star-v-sagittae-should-go-nova-within-this-century/) "In 1902 astronomers discovered V Sge as a variable star, and in 1963 a team of astronomers led by American George Herbig identified it as a binary star." Yet the technical details pdf says "V Sge was discovered in the year 1904, and there have been visual magnitude estimates continuously since then." Sure, the date might be mostly trivial, but these facts are essentially just inaccurate and thus interpretative (a source of errors.)
I have no disagreement with V Sge own uniqueness, but there are shared circumstances with other similar systems. Interpretation by some have implied a new kind of astrophysical phenomena, which is not necessarily true. e.g. In saying: "With all other CVs being in a completely different regime of mass transfer, all common experience and theory from the CV community is completely inapplicable to V Sge." (Its uniqueness is, of course, being near the very end of the system's evolutionary process, but it more importantly explains why novae are such rare events. Your technical paper does explain why: "V Sge is the only known Cataclysmic Variable (CV) that has the q>1 case. All other CVs have q<1, and even the few so-called ‘V Sge stars’ all have q<0.5", but the nuance isn't so clear in the media not fully understanding 'q'.)
In all, I see this problem as some kind of a communication issue, essentially explaining something very complex so novices can understand and digest the information. The media regularly end up going or grabbing for the greatest impacting outcome without using any necessary caution, often overplaying the research just to make it more interesting. Historically there have been many direct examples. e.g. Predicting 2012's sungrazing Comet Ison reaching −12 magnitude or presumably discovering that a LMC cluster 30 Doradus R136 containing several +300 mass stars (R136a1) So why do some articles somehow became viral in the media? The problem IMO is the disconnect between the researcher(s) expressing their work and knowledge and the ability of the media reducing that into simpler words. As with all new discoveries and declarations the truth will be revealed in hindsight (and not at the time of announcement). My admonishment is only against the degree of imprecision here.
Regardless of any of this, I do look forward in reading the formal paper and the S&T publication!
Q: Is it at all possible the nova TNR ignition could happen before the predicted merge event in 2083? If that did happen, would this not extended the time before the stellar remnants do finally merge, and consequently, also end this runaway behaviour? (as Don-Brundage comment pointed out.)
Note: The new V Sge chart looks great.
You must be logged in to post a comment.
Brad-Schaefer
January 19, 2020 at 9:24 pm
Yes, it is difficult to accurately express complex results in a short popular article for general consumption. For astronomy and science, that is why good science journalists are so valuable. Indeed, with the importance of a wide range of science results with large societal impact, and groups actively trying to deny strong science results, I consider a good science journalist as more important than a good scientist. For letting people know about fun and exciting astro-results, with the widespread availability of poor-or-bad astro-claims throughout the internet and such, good astro-journalists are our best source for news and our best defense. We gotta give a lot of credit to the good journalists for their great help for our field.
For Sky & Telescope in particular, a few years ago, I performed a study of many astronomy articles in the popular media, specifically aimed at the twin questions of "Is the reporter telling the basic story correctly?" and "How reliable is the astronomy discovery reported in the media?" So I had 377 articles reviewed by each of three world-class experts (including a Nobel Prize winner). These articles appeared in each of five sources; Sky & Telescope, the New York Times, UPI, Science News, and four middle-sized city newspapers. The 377 articles were all found in those sources on any of three topics that broadly span modern astronomy (supernovae, Mars, and gamma-ray astrophysics) for which I could line up the experts. From these reviews, I found that the Sky & Telescope reporters never made errors, big or small, in any of the sampled articles. (The same for Science News.) This is why I have confidence in S&T, and why this is even now my dominant source of astro-news, and has been since junior high school. For the newspapers with their much faster deadlines (often daily), I was expecting a poor showing. (Indeed, many studies by journalist researchers show that scientists have quantifiably poor expectations for newspapers.) I was startled to find that the newspapers only had an average of 0.5 small-errors/article, for issues that did not affecting anything central or important). And for significant errors (anything that affected any real understanding), we found an error rate of 0.00 large-errors/article. Wow, even daily-deadline reporters are doing a generally very good job. We scientists and the general public should realize and respect the daily newspaper reporters as being reliable and worthy.
So the science journalists should receive our respect and thanks for their difficult jobs. But how can we reconcile this conclusion with both our realizations that we've seen much bad reporting on the internet. I think that this reconciliation comes in two parts: First, the regular science journalists for established news media outlets are professionals of top quality, while the bad internet reports come from often-private 'enthusiasts' of little-knowledge and irrelevant axes-to-grind. This is the reason why good science journalists on reputable media outlets are so valuable. Second, too often, especially in recent decades, the science story is forced into a very short text, either by a disinterested editor or by the format of the internet web page, and, as you commented, it is hard to get the story right or useful when it is crammed into too few words. Both of these problems for astronomy journalism are also big problems for many other fields, as well as for larger societal problems. I have no good solution for how to rid the world of poor articles, the crackpots, and the trolls. Nevertheless, for astronomy journalism, Sky & Telescope is a substantial part of the solution.
You must be logged in to post a comment.
Brad-Schaefer
January 19, 2020 at 10:05 pm
At the end of your note, you asked a detailed technical question, about the possibility of a classical nova eruption on V Sge before the merge-burst sometime in the decades around 2083. (This merge-burst will have a total energy somewhere between those of novae and supernovae, while the fast eruptive light curve with a ~month-long peak and a slow decay has good similarities to the light curves of novae and supernovae.) The reason to possibly expect a classical nova event is because V Sge is a Cataclysmic Variable, where gases from the companion star are spilling onto the white dwarf, and we might expect a thermonuclear runaway when enough mass piles up. Alas, but this ordinary expectation is denied because V Sge uniquely has a humongous rate of mass spilling over onto the white dwarf. (V Sge has an accretion rate roughly a factor of a hundred more that any other Cataclysmic Variable.) Theorists have long and confidently known that no nova eruption will occur if the accretion rate exceeds some critical threshold. From the classic paper of Ken'ichi Nomoto (1982, ApJ, 253, p.798, see Figure 9), the 0.85-M_sun white dwarf can suffer a nova eruption only if the accretion rate is less than 10^-7 solar masses per year. V Sge has >10^-5 solar masses transferring each year, so instead of a nova flash burning the hydrogen, the gas will have on-going steady hydrogen burning and the material will puff up into a red giant configuration around the white dwarf and even drive a fast wind. So there is no chance of V Sge 'going nova'.
Predictions have a critical and key role in science. In principle, science works by making predictions from some model or theory, and then going out and testing those predictions. This is a never ending cycle, where a model or theory that repeatedly and successfully passes its predictions through the filter of reality comes to be more-and-more confidently accepted. Such predictions must be testable. For V Sge, we are predicting a specific brightening over the next decade, and we are predicting a spectacular merge burst round about the year 2083. Our predictions are testable on both a moderate and a long time scale. So, the V Sge predictions are just a normal part of ordinary science, part of the endless cycle of making and then testing predictions.
You must be logged in to post a comment.
John
January 11, 2020 at 4:03 pm
At 140, I think I'll give it pass, maybe see it from the other side...
You must be logged in to post a comment.
You must be logged in to post a comment.