About the SA Blog Network

Life, Unbounded

Life, Unbounded

Discussion and news about planets, exoplanets, and astrobiology
Life, Unbounded Home

The Stars Are Beginning To Go Out…

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

The galaxy NGC 1365 aglow with H-alpha light that tends to show star forming zones(Credit: ESO)

They really are.

The universe is apparently well past its prime in terms of making stars, and what new ones are being made now across the cosmos will never amount to more than a few percent on top of the numbers already come and gone.

This is the rather disquieting conclusion of a new and significant study of the rate at which stars have been produced through cosmic time.

Sobral and colleagues recently published the results of a series of ‘snapshots’ made of galaxies busily making stars at different epochs, from about 4 billion years ago (around the time of Earth’s formation) all the way back to nearly 11 billion years ago. This is no simple task, some of the world’s largest and most sensitive telescopes had to be employed.

By observing light at very specific frequencies (corresponding to emission from warm hydrogen atoms – see the note below) they are able to gauge the actual rate at which new stars are condensing out of thick nebular material in a few thousand galactic systems. This yields some very robust statistics on the global changes in the numbers of new stars being made as the universe ages.

The main conclusions come in two parts. First, 95% of all the stars we see around us today were formed during the past 11 billion years, and about half of these were formed between roughly 11 and 8 billion years ago in a flurry of activity. But the real shocker is that the rate at which new stars are being produced in galaxies today is barely 3% of the rate back 11 billion years ago, and declining. This indicates that unless our universe finds a second wind (which is unlikely) it will only ever manage to produce about 5% more stars than exist at this very moment.

This is, quite literally, the beginning of the end.

However, despite the provocative title of this post you shouldn’t actually expect to see the stars start to disappear from view too soon. The great majority of stars in the universe are less massive than the Sun, and in the Milky Way about 75% of all stars are less than half as massive. Smaller stars last longer – the nuclear fusion of hydrogen happens slower in their cores and they also tend to churn their innards more than a star like the Sun, resulting in some extraordinary efficiency. Indeed, the smallest stars (so-called M-dwarfs) should have trillion-year lifetimes.

So we live at an interesting time, at the cusp between exuberant excess and a long gentle decline. We also happen to live in a galaxy that still produces a few stars a year – the Milky Way is going to end up contributing nicely to that last 5%. And when the Andromeda galaxy comes lumbering into us in 4 or 5 billion years time there may be a sudden burst of new star formation as these two beasts merge, and a final sprinkling of new stellar beacons will – for a time – light the cosmos a little more.

[As a small side note, a number of otherwise respectable media outlets - like this - mistakenly reported that the astronomers performing this study had been detecting 'alpha particles emitted by Hydrogen atoms'. This is such a heinous error that I feel compelled to mention it.

First of all, no hydrogen atom in the history of the universe has ever emitted an alpha-particle since an alpha-particle consists of two protons and two neutrons - in other words a helium nucleus.

Second, the astronomers were actually just measuring light emitted in the hydrogen-alpha atomic transition, when an electron drops from the third major allowed energy level around a proton to the second, and spits out a reddish photon. These reddish photons are shifted to lower and lower energies or frequencies as they pass through the expanding universe, and for the really distant objects studied in this case they end up in the infrared part of the spectrum. But they're a nifty way to pin down regions in galaxies where stars are forming and the H-alpha photons penetrate through gas and dust quite well, so can escape to eventually be seen by the likes of us.]

Caleb A. Scharf About the Author: Caleb Scharf is the director of Columbia University's multidisciplinary Astrobiology Center. He has worked in the fields of observational cosmology, X-ray astronomy, and more recently exoplanetary science. His books include Gravity's Engines (2012) and The Copernicus Complex (2014) (both from Scientific American / Farrar, Straus and Giroux.) Follow on Twitter @caleb_scharf.

The views expressed are those of the author and are not necessarily those of Scientific American.

Previous: The Hot Baths of Mars More
Life, Unbounded
Next: Mars Tinted Goggles

Rights & Permissions

Comments 21 Comments

Add Comment
  1. 1. rloldershaw 10:47 am 11/19/2012

    C.S.: “This is, quite literally, the beginning of the end.”

    Unless the spacetime of the cosmos is infinite and nature is an eternal hierarchical system.

    In such a cosmological paradigm objects at all Scales of the infinite hierarchy are being excited and de-excited, and are annihilated or forming, in comparable numbers when the entire infinite cosmos is taken into account.

    Thus the entire cosmos is in eternal balance. Even if parts of it “wink out” there are other places where new subsystems are forming.

    Robert L. Oldershaw
    Discrete Scale Relativity

    “How can physics live up to its true greatness except by a new revolution in outlook which dwarfs all its past revolutions? And when it comes, will we not say to each other, ‘Oh, how beautiful and simple it all is! How could we ever have missed it for so long!’.” John Archibald Wheeler

    Link to this
  2. 2. lamorpa 11:16 am 11/19/2012

    And only 6,000 years old!

    Link to this
  3. 3. littleredtop 12:39 pm 11/19/2012

    Apparently, the “Henny Penny” theorists are unaware that matter doesn’t just come from nowhere. There is only a given amount of matter in our Universe and it is constantly being redistributed in small quantities as solar systems implode and new stars and solar systems are formed. Our solar system is still in its warming stage (global warming) and relatively new. I suspect that it won’t be for quite some time before the “Henny Penny” prophesy is fulfilled here on Earth.

    Link to this
  4. 4. jtdwyer 1:47 pm 11/19/2012

    “These reddish photons are shifted to lower and lower energies or frequencies as they pass through the expanding universe, and for the really distant objects studied in this case they end up in the infrared part of the spectrum.”

    Aren’t the more distant of these photons now shifted to the infrared spectrum more capable of penetrating gas & dust than nearer ‘reddish’ visible light spectrum photons, potentially producing some observational bias?

    Link to this
  5. 5. MadScientist72 2:13 pm 11/19/2012

    How do we know that we’re not simply in a slow corner of the universe? The data we’ve got for our immediate galactic neighborhood is all relatively recent, but the data sets get progressively older the further out you get. Once you reach the farthest extent of what we can observe, we can see conditions ~13.75 billion years ago, but nothing more recent than that. That’d be like trying to assess global population trends when you only have current stats for a few square miles of Iowa and only have data from the Stone Age by the time you get to Australia.

    Link to this
  6. 6. Caleb A. Scharf in reply to Caleb A. Scharf 2:46 pm 11/19/2012

    jtdwyer – good qn. but in general there is little between us and the galaxies where these photons originated, so the bias effect is very small.

    Link to this
  7. 7. Caleb A. Scharf in reply to Caleb A. Scharf 2:51 pm 11/19/2012

    MadScientist72 – you’re right that the data in this case (and in most of this type of survey) spans an enormous range of space and time, and about 3,000-4,000 individual galaxies spread over that range (in 4 main ‘time slices’ or redshift ranges to be exact). In our ‘local’ universe we actually have even more data, hundreds of thousands of galaxies in effect, from which the local star formation rate can be evaluated in a number of ways. So the real challenge – that seems to have been met by this latest study – is gathering enough of the distant and ancient data rather than local data.

    Link to this
  8. 8. EyesWideOpen 4:19 pm 11/19/2012

    What if it’s the other way around? Suppose it’s not the stars that are “going out” (through reduced star formation).

    What if we are “going out”? What if our segment of the universe is ceasing to exist through some celestial event on a macro scale we cannot measure from this world — creating the illusion the universe in our periphery is in decline.

    Link to this
  9. 9. jtdwyer 4:35 pm 11/19/2012

    Caleb A. Scharf – Thanks, but the constant filter that all intergalactic light must pass through is the gas & dust clouds within the Milky Way. Unless clouds of gas & dust were avoided in this survey, mightn’t there have been some observational bias based on the varying spectra selected?

    Link to this
  10. 10. caleb_scharf 4:44 pm 11/19/2012

    jtdwyer – yes, you’re quite correct, but we have very detailed physical maps of all this Milky Way material and it is routine to apply so-called ‘extinction corrections’ to all extragalactic observations to account for these effects. Also, in this case the survey basically employs narrow-band filters tuned to particular redshift (distance) H-alpha emission, so treating the extinction corrections is not too complex.

    Link to this
  11. 11. SteveO 6:05 pm 11/19/2012


    …and apparently you don’t believe in the laws of thermodynamics. I have a perpetual motion machine to sell you!

    Still, it will be a few billions of years before our sun poops out. But poop out it will…

    Link to this
  12. 12. jtdwyer 6:30 pm 11/19/2012

    caleb_scharf – Thanks for explaining!

    Link to this
  13. 13. MadScientist72 8:39 am 11/20/2012

    @ Caleb A. Scharf – No mater how much “distant and ancient data” we collect, it won’t change the fact that distant galaxies will only be able to provide us with ancient data and nearby galaxies will only be able to give recent data, so we’ll never be able to be sure that our nearby galaxies are typical of the universe as a whole. We could be in a region of galactic anomalies that where star formation has always slower than normal. Unless you’ve got a few billion years to wait for the light that’s currently leaving those distant galaxies to get here, there’s no way to compare near & far at the same timepoint.

    Link to this
  14. 14. caleb_scharf 9:21 am 11/20/2012

    MadScientist72 – well, by that logic we might as well not bother doing any astronomy. I strongly disagree that we can’t be sure we’re not in a region of ‘anomaly’. While the study in this post makes use of what are effectively a set of snapshots or slices through cosmic time we now have surveys that map *continuous* regions of the universe from our location out to hundreds of millions (billions) of light years away – and there is no apparent discontinuity. Yes, we’re always seeing ancient stuff, but the statistics of the populations (galaxies, stars, supernova etc etc) show smooth and coherent evolution with cosmic time – i.e. it is not hard to see how other regions of the universe will end up precisely like our own. There is also no confirmed evidence (although there are a few claims) of any measurable variations in fundamental physical constants between entirely separate regions of the universe.

    Now, I would agree that on *really* large scales, beyond the present horizon, we don’t know whether things are different, but that’s beyond the observable universe.

    Link to this
  15. 15. Knyaz 7:53 am 11/21/2012

    В нашей вселенной всё взаимосвязано поэтому если в одном месте что то происходит то это обязательно что хоть немного но меняет в всей вселенной.Например экстремальное изменение погоды это показатель изменения климата,изменение климата это показатель изменения альбедо Земли (антропогенный фактор и увеличение количества метана это ускорители измения климата).Изменение альбедо (от альбедо а не от человека зависит количество солнечных лучей на Земле) и деформация Земной коры это показатель измения формы Земли,измение формы Земли это показатель изменения процессов в ядре Земли.Дальнейшее изменение формы Земли изменит скорость вращения Земли вокруг своей оси,угловой наклон,что нарушит равновесие в системе Земля-Луна.Вселенная это система поэтому любые изменения в любой части этой системы обязательно нарушат равновесие в системе.

    Link to this
  16. 16. Wayne Williamson 3:42 pm 11/23/2012

    Are you sure you posted to the correct article.
    I don’t know if this had anything to do with global warming or any other changes on the earth(or the shape of the earth)…

    Then again, maybe google translate didn’t do a good job of translating your post…

    “In our universe, everything is interconnected so if one place that it is necessary that it is a little bit but I was in the whole vselennoy.Naprimer extreme weather change is an indicator of climate change, climate change is a measure of change in the Earth’s albedo (human factor and increase the amount of methane is boosters climate change). changing the albedo (the albedo and not of man depends on the amount of sunlight on the Earth) and the deformation of the Earth’s crust is a measure of change the shape of the earth, change the shape of the Earth is a measure of the change in the core processes Zemli.Dalneyshee change in shape of the Earth will change the speed of rotation of the Earth its axis angles that upset the balance in the Earth-Luna.Vselennaya a system so that any changes in any part of the system necessarily disrupt the balance in the system.”

    Link to this
  17. 17. ABJ 6:31 pm 11/23/2012

    In the “Nine BIllion Names of God” Arthur C Clarke wrote about two computer technicians hired by a Nepalese temple to have the computer print out the 9 billion names of God. After completing the programming, the technicians fear what might befall them if nothing happens when the computer finishes the task. They slip out just before the job is done.

    Under a clear night sky they estimate that it must be just about the time that the monks are pasting the final printed names into their holy books. Then they notice that “overhead, without any fuss, the stars were going out.”

    Link to this
  18. 18. Quinn the Eskimo 1:24 pm 11/25/2012

    Replace them with CFL’s. They’re more economical. More expensive. And don’t last as long.

    Financial Stimulus!!!

    Link to this
  19. 19. vinodkumarsehgal 11:17 am 11/26/2012

    Article states that current rate of formation of stars is 3% of what it was 11 billion years ago. However, it does not indicates the reasons for the rapid reduction in the rate of formation of stars. Is it due to lack of raw material in the form of gas in nebulas Or due to lack of “right conditions” for the formation of stars or something else? Any how, have astronomers pinned down and constrained specifics of the right conditions for the formation of stars. If it is so, do lack of specifics of “right conditions” match with the prevailing conditions within galaxies especially MW galaxy where rate of formation of stars has found to be very low.

    Link to this
  20. 20. vinodkumarsehgal 11:51 am 11/26/2012

    CS : “but we have very detailed physical maps of all this Milky Way material and it is routine to apply so-called ‘extinction corrections’ to all extragalactic observations to account for these effects”

    In case of MW galaxy, it may be possible to apply “extinction corrections” due to availability of detailed physical map. But what about ” extinction correction” for innumerable galaxies en-route the light emitted from galaxies billion of light years away. Astronomers are not sure about the galaxies which lie on the route of light as well as the physical conditions prevailing in the galaxies at the time of transit of light It will be quite naive to assume that light from remote galaxies emitted billion of years ago enters directly to MW without passing thru other galaxies. Astronomers estimate about 100-200 billion galaxies ( within their observational frame of observable universe only), as such, there should be no wonder that light from remote galaxy located at distances as large as 10-11 billion light years, may pass thru some million of galaxies and get affected by gas/dust prevailing in those galaxies. In an expanding universe, conditions of universe on the trajectory of emitted light in terms of galaxies and conditions of dust/gas within those galaxies may have a dynamic change every moment and whole scenario may emerge out to a quite complex one. Further, In an expanding universe, how to ascertain that light maintains its straight trajectory for an unfathomable distance running into some 10-11 billion light years?

    In view of above, how to apply the “extinction correction” for the emitted light in totality?

    Link to this
  21. 21. Maredeistelle 2:30 pm 01/10/2013

    But if everything is made by energy, and energy cannot be destroyed, doesn’t that mean that the Universe would never be destroyed?
    And also, is there any chance that this is just a passage to a cosmic era of steadiness? Maybe the Universe will just have the same amount of stars for billions of years or something like that.

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Email this Article