Someone went and got me a subscription to Scientific American, so for the past few months I’ve been covering physics articles, now with the benefit of a PhD. Perhaps it’s a way to keep in touch with my physics roots as my career has moved on to other things.
In this month’s issue, the cover article is “A Cosmic Crisis“, about a discrepancy between two measurements of the age of the universe.
Funny thing, there’s always a letter from the editor in chief where he introduces all the major articles, and here he contrasts the “cosmic crisis” with another ongoing crisis, that thing between US and Iran. Yep, this sure is an article that was written last month! FWIW, I could do with some reading that has nothing to do with COVID-19.
I thought I’d review the article in approximately the order in which I read it: pretty pictures first, walls of text last if at all. For serious, this is the correct way to read a science article, I can say that as a person with a PhD.
Credit: Scientific American
There’s one picture that immediately grabs me, which is a plot of Hubble Constant Estimates vs time of publication. It shows two measurements, one from “Distance ladder method using Cepheids” and one from “CMB measurements”. These measurements have gotten more precise over time, and starting in 2014 (which by scientific standards is very recent) the two estimates diverged. Right now it seems they’re about 3 standard deviations apart. There’s also a newer third measurement, “Distance ladder method using flaring red giant stars”, which now seems to disagree with each of the other two.
From this one plot, I can already pretty much tell what the article is about, because I have a bit of prior understanding of the Hubble Constant and the methods used to measure it.
- The Hubble Constant is a number that measures the current rate of expansion of the universe. Contrary to the name, it’s not actually a constant, the Hubble Constant slowly decreases over time according to general relativity. The standard model that predicts how the Hubble Constant changes over time is called the ΛCDM model. By extrapolating backwards you can figure out the age of the universe.
- The CMB is the cosmic microwave background, which is the the very oldest light signal we can observe in the universe, because it comes from the time when the universe became transparent. Small variations in the CMB in different parts of the sky are basically quantum fluctuations writ large, and somehow this can be used to measure the Hubble constant.
- I’m not exactly sure what Cepheids are, but I know they’re a “standard candle”, an object that shines with predictable luminosity (i.e. total amount of light power). If you know the luminosity, and how bright it appears from Earth, you can tell how far away they are. And their redshift can be used to measure how fast they’re receding. So that gives you the Hubble Constant.
- I’ve never heard of the flaring red giant method before, but I surmise that astronomers found another standard candle.
Where could these methods be going wrong? I’m sure a lot of astrophysicists have thought really hard about it!
So let’s move on to the text of the article. By the way, the normal reading order for the text of a paper is: abstract, introduction, conclusion, discussion, methods. But since this is a popular science article I guess they expect me to read it in order. Anyway, here’s what I learned:
- Red giants are a standard candle because dying red giants undergo a “helium flash”. The internet tells me that’s when the helium at its core starts fusing. Why does the helium flash have predictable luminosity? I have no idea. In fact, I don’t understand red giants at all. Grade school science taught me that red giants form when stars run out of hydrogen fuel, but it takes a degree in physics to realize that I don’t have any idea why that is.
Uh, well that’s about it. There were a lot of historical details, and an overarching narrative about astronomers (who make the standard candle measurements) vs cosmologists (who predict the Hubble constant from CMB measurements). But no real solutions, not even proposed solutions. I feel like I got more out of the one graph than the rest of the article. Well, it’s a really good graph!
Anyway, sounds exciting for the researchers in this field. I’m sure cosmologists are rubbing their hands over the thought of overturning the ΛCDM model.
Thank you, thank you, thank you for saying this. That happens to be my opinion — speaking as a mathematician and engineer (as well as a generally-interested lay-person) — as well. I find doing so not only tends to increase my interest in “the details” (wall(s) of text), but also “preps” me for what the text is saying / explaining. Of course, there are downsides: As one example, poorly done illustrations / tables / whatever can be seriously misleading (i.e., that “prep” is “too wrong”), or can “turn off” my interest in something that is really interesting.
Also, thanks for the summary of the article. I’m in Covid-19 lockdown here in France, which among other things, means I shouldn’t go out to buy a newspaper, &tc, including SA. (I don’t have a subscription, I gave it up years ago for reasons I don’t now clearly recall.)
In my experience, figures and tables aren’t just the part you read first, they’re also what’s created first! Generally, they’re the most important part of a (physics) paper, and should tell most of the story on their own. A poorly done figure/table does not bode well for the rest of the article.
@2, Good point. Thinking back, that is quite probably the case in some my own work (I cannot speak for others), albeit it tends to be an iterative cycle: Diagram (rough sketch or whatever), text, revise diagram, revise text, and so on, until one is happy the point is being explained. Then it’s reviewed and one discovers it isn’t being explained, so sortof-start over again… 😉
Siggy @2: Not generally the case for theoretical papers.
@Rob Grigjanis,
True, although a lot of theory papers I read had figures from numerical simulations. The others… well the main thing I learned was how to give up quickly.
On the side, I’ve done lots of different graphics and such for music theory papers — various forms of music notation too, which is how I got into it (since I write the stuff), but plenty of diagrams and charts and tables and all kinds of stuff really. There’s not enough pretty pictures in those either, if you ask me. I’d say that, for anyone writing an academic paper, no matter what the field is, it is really helpful to find a person who understands the content, rather than someone who might be a good graphic designer but is otherwise clueless. It can save a lot of time and headaches, but it’s really just as simple as “two heads are better than one.”
And it is interesting how much visuals can help to not only communicate the ideas to other but can also help to formulate them in the first place. I remember a bunch of times saying things like this (as politely as I can):
Once you actually see it in front of you, or while you’re in the process of figuring out how to show it — but sometimes, it simply can’t be represented, which is typically a bad sign — that can definitely help to clear up some of the cobwebs. So you just have to reassess and go back to the proverbial drawing board.
In one example that comes to mind, I threw together some graphics purely to see what might come out of it, since the analysis at the time just wasn’t very clear (or I didn’t feel like I had a good enough grasp of it). Maybe it was dumb luck, but it was surprisingly useful. Not quite a “Eureka” moment all on its own, but the approach after that did shift pretty dramatically in that direction, since it was hard to ignore what we were looking at.