DES and the BAO Scale
I just saw a press release about new results from the Dark Energy Survey relating to measurements of baryon acoustic oscillations. These are basically the residue of the oscillations seen in the power spectrum of the cosmic microwave background (CMB) temperature distribution imprinted on the galaxy distribution. They are somewhat less obvious that the primordial temperature fluctuations because the growth of structure produces a much larger background but they are measurable (and indeed are one of the things Euclid will measure).
Anyway, there is a very nice detailed description in the press release and you can find the preprint of the work in full on arXiv here, so I’ll just show the key figure:
The effective redshift of this measurement is about 0.85; in the CMB the redshift is about 1000. You can see that there is a characteristic scale but it is slightly offset from that predicted using the standard ΛCDM model based on the Planck determination of cosmological parameters. One has to be careful in interpreting this diagram because it is determined using autocorrelation functions; the errors on different bins are therefore correlated, not statistically independent. They are also, as you can see, quite large. Nonetheless, it’s a tantalizing result…
In the Dark 
February 23, 2024 at 8:18 pm
Assuming that Planck’s determinations of cosmological parameters have nonzero margins of error (which seems a reasonable assumption), the “Planck template” represented by the orange dotted line should have error bars of its own. In that case, the discrepancy between the Planck LCDM prediction and the DES Y6 measurement of the angular BAO would, I think, be greatly reduced.
February 24, 2024 at 7:23 am
The uncertainty in the standard model prediction is probably taken into account in arriving at the conclusion. The best curve only is shown on the plot, presumably because it would luck cluttered if festooned with error bars.
February 24, 2024 at 9:52 pm
I’m afraid I don’t see where the uncertainty in the Planck LCDM prediction is taken into account. On p. 30 of the arXiv preprint (paragraph 1 of section VIII.A), the Planck LCDM value for the ratio of angular diameter distance to acoustic scale is given simply as 20.39, without specifying any margin of error. (The DES value, in contrast, does come with a nonzero margin of error.) This is consistent with the lack of error bars for the “Planck template” depicted in the graph in question. I suggest that including a margin of error term in the Planck LCDM prediction would entail some modification of DES’s conclusion (on p. 30 of the arXiv preprint) that their measured BAO value differs from Planck’s by 2.1 sigma; the revised difference would probably be less than 2 sigma.
February 25, 2024 at 1:42 am
You may be right. Why don’t you email the lead author and ask?
February 25, 2024 at 4:00 am
Good idea- thanks for the suggestion!