Uncommon Descent Serving The Intelligent Design Community

Experimental physicist Rob Sheldon takes issue with News’s globular clusters story

binary star evolution within a globular cluster/Mark A. Garlick, U Warwick

Recently, we published a story on a recent research finding that globular clusters may be four billion years younger than previously thought. In a universe that is only a little over 13 billion years old (so far as we know) that seems to some of us like a large room for error.

However, Rob Sheldon writes to say,

This is actually a good thing, not a bad thing.


From astronomy 101, we talk about the Hertzsprung-Russell diagram. HR diagrams plot the stars in our galaxy on a 2D plot, where the vertical axis is luminosity or brightness, and the horizontal axis is temperature or color. Most stars are clustered around a straight line, with blue-white giants stars like Rigel in the upper left, and red dwarf stars in the lower right. The clustering means that stars spend most of their life on the “main sequence”. With the development of Los Alamos “bomb codes” in the 1950s, we learned that the main sequence describes stars that are burning (fusing) hydrogen. These stars will burn at an increasingly fast rate proportional to their size–blue-white stars are 300 times bigger than red dwarfs. But they burn out faster too, blue-white lasting 100My, whereas red dwarf will last 30,000My. When they run out of hydrogen, they start to burn helium, but it isn’t as good a fuel, and (if big enough) they become red-giant stars like Betelgeuse. These red giants aren’t on the “main sequence” but off to the top-right corner of HR diagram known as the “red giant branch”.


If we observe a globular cluster and plot all the stars on an HR diagram, you can see where the bigger stars have left the main sequence and joined the red-giant branch. It looks like a sharp corner in the HR diagram. This turn-off point gives you the age of the globular cluster–assuming all the stars were born together.

Only some clusters were older than the age of the universe!

This paper shows that the bomb-codes forgot to include the effect of binary stars. Having a very close companion will “age” stars faster than usual because they “steal” the outer envelope of gas, causing it to run out hydrogen too soon. When this extra “aging” is taken into account in the HR diagram, the clusters are actually quite a bit younger than they thought, and the Milky Way’s clusters are not 12by but only 9by old, which is a good thing, because otherwise the clusters are older than the galaxy.


A physics model is always incomplete. Some missing piece of trivia will turn out to be important. So always take models, even elaborate expensive super-computer models, with many grains of salt. They are only as good as the assumptions that go into them.

Actually, I (O’Leary for News) don’t think it is a bad thing. I think lectern-splintering is a bad thing. I can easily live with the fact of admitted uncertainty. The pretense of certainty – all too evident in many science venues – gives one pause for thought. Four billion years plus or minus is a lot of uncertainty in the context.

Note: Rob Sheldon is the author of Genesis: The Long Ascent

See also: Are globular clusters 4 billion years younger than previously thought? It “brings into question” more than the mechanics of galaxy formation. There is considerable distance between nine billion years and thirteen billion years. An equivalent claim for life on Earth would shave a billion years off the development of life. If it’s true, it’s true. But the finding doesn’t fill onlookers with confidence about the accuracy of dating systems.

Sheldon's point answers my concern on age index by branch point. Multiple stars are common, so it is plausible that binaries -- which are gravitationally coupled -- would have a material effect. KF kairosfocus

Leave a Reply