≡ Menu

Star Formation in the Early Universe

We know that new stars form out of cold gas and dust that are present in galaxies, but what accounts for the fact that star formation is slower than in earlier eras? Three to five billion years after the Big Bang, galaxies turned out stars at a much faster clip than they do today. The Milky Way seems to produce stars at a rate equaling about ten times the mass of our Sun each year, whereas similar galaxies earlier in their lives featured star formation rates that were up to ten times higher.

Michael Cooper (University of Arizona) and colleagues have gone to work on this question by studying data from the DEEP2 survey of 50,000 galaxies, picking a sample of one dozen massive galaxies to represent the average population. Working with the Hubble and Spitzer space telescopes as well as radio telescope arrays in France and California, the team then observed the selected galaxies in the infrared and measured their radio frequency emissions, making cold gas clouds visible.

Cooper and company were trying to discover whether older galaxies had a greater supply of gas and dust than similar galaxies today, or whether the efficiency of star formation somehow drops with time. This is tricky work because the dense, cool clouds of molecular gas in which star formation will occur are found at temperatures between 10 K and 100 K, emitting only tiny amounts of visible light. Having coalesced, a star’s radiation then dispels the gas and the star is visible.

Image: Viewed through the Hubble Space Telescope at visible light (left), a galaxy does not reveal its full secret underlying star formation. Only when observed using a combination of radio emission and infrared wavelengths, the galaxy reveals a massive, rotating disc measuring about 60,000 light years across (right). This disc consists of cold molecular gas and dust, the raw materials from which stars are born. Credit: University of Arizona.

The results seem straightforward, according to co-author Benjamin Weiner:

“What we found now is that galaxies like the ancestors of the Milky Way had a much greater supply of gas than the Milky Way does today. Thus, they have been making stars according to the same laws of physics, but more of them in a given time because they had a greater supply of material.”

Previous studies of this question had focused on bright objects, galaxies that were easier to study but not necessarily representative of the broader galactic population. Cooper’s team refined the methodology to select galaxies that could be considered ‘normal,’ firming up our picture of how galaxies make stars. They chose galaxies at redshifts of approximately 1.2 and 2.3, when the universe was 40% and 24% of its current age.

Adds Cooper:

“From our study, we now know that typical galaxies in the early universe contained three to ten times more molecular gas than today, a strong indication that the rate of star formation has slowed because those galaxies have less raw material available compared to when they were younger, and not because there was some change in efficiency with which they make new stars.”

The paper is Cooper et al., “High molecular gas fractions in normal massive star-forming galaxies in the young universe,” Nature 463 (11 February, 2010), pp. 781-784 (abstract).


Comments on this entry are closed.

  • Doug M. February 17, 2010, 12:06

    This raises an interesting question: in the future, will star formation in the Milky Way gradually taper off along some sort of (exponential? power) curve?

    Or will it stop more abruptly, in a nonlinear way, when available material drops below some critical density?

    Doug M.

  • Ronald February 18, 2010, 5:51

    I basically had the same question as Doug (though I presume that the abrupt ending of star formation would be the result of a linear decline, not non-linear, which would rather result in mentioned gradual tapering-off).

    And logically related to this: I used to think that the hydrogen gas reserves of the universe are still huge and just a small dent has been made in those by star formation so far. Apparently not? Or is this, again, related to some threshold density, below which star formation drops off sharply.

    If this is indeed the case, then the stelliferous life expectancy (now usually mentioned as on the order of 100 thousand gigayears) could be considerably lower.

    More light on this anyone?

  • Ronald February 18, 2010, 5:58

    And another related question suddenly comes up: with decreasing stellar formation rate, is there also a (progressing) change in the *types* of stars that are formed?
    E.g. relatively more smaller stars (red dwarfs), and fewer large ones. I think I read something like this somewhere, that, as the universe progresses toward say a 100 gy, formation of stars larger than M dwarfs and late K will become very rare (including sunlike star formation).

  • Adam February 18, 2010, 7:48

    Hi Doug M.

    A good book to look at for your answers is Fred Adams & Greg Laughlin’s “The Five Ages of the Universe”. They conclude that star formation will get slower and the ignition mass of fusion-powered stars will get lower as the average metallicity goes up. Eventually not enough fusion fuel is left and molecular clouds are too full of ‘metals’ to collapse in a form heavy enough for fusion. Thus after about 100 trillion years star formation will have ground to a halt.

  • keith February 23, 2010, 8:56

    I’ve recently got hooked on the publications of Rudolph Schild and Carl Gibson. they contend that theories of unniverse structure formation, including star formation, are fatally flawed. This is because of the oversimplifications of the Jeans calculation.

    They contend that dark matter is baryonic, and consists of planetary mass H/He objects in clusters. There are about 30 million per visible star. Stars are formed when these clusters merge.

    Everything is explained, the need for dark energy evaporates, once you realise the achromatic fog of planetary bodies is causing greater than expected optical density towards objects at high redshift. Supervoids also fall naturally out of the equations.



  • forrest noble March 2, 2010, 19:55

    “….but what accounts for the fact that star formation is slower than in earlier eras? Three to five billion years after the Big Bang, galaxies turned out stars at a much faster clip than they do today.”

    I think in time it will be realized that this assessment is wrong. These are conclusions based upon the perspective that from this distance we can properly asses the rate of galactic star formation which includes a number of assumptions, some of which I believe will also turn out to be wrong.