In March of 2024 the DESI collaboration dropped a bombshell on the cosmological community: slim but significant evidence that dark energy might be getting weaker with time. This was a stunning result delivered after years of painstaking analysis. It’s not a bullet-proof result, but it doesn’t have to be to make our lives more interesting.

I know I’m late to the party on discussing this. And it’s okay, because 1) there’s a lot to unpack in this kind of result and I wanted to take my time, and 2) it’s not like this result is going to get revised or even updated anytime soon, so we’ve got plenty of room to play with this.

Let’s start with the results themselves and how they got there. DESI stands for Dark Energy Spectroscopic Instrument. It’s a roughly 4-meter telescope mounted on Kitt Peak in southeastern Arizona. It’s a galaxy survey, and to accomplish this survey they have 5,000 robotically controlled fiber optic cables underneath the telescope. Every night, the telescope selects a patch of sky to observe, the robots position the fiber optic cables to align with the positions of galaxies within that patch, and the instrument records detailed information for each and every single one. Then they do the same thing the next night, and then the next, and then the next.

So far they have amassed a catalog of over 13 million galaxies, providing the largest and comprehensive survey of galaxy positions in history. And they’re not even done! They’re aiming for 50 million galaxies once the survey is complete.

And let me tell you, those robotically controlled fiber optic cables are a huge game changer. In many ways DESI is the successor to an older survey, the Sloan Digital Sky Survey. That survey had a similar setup, except that instead of robots to move all those fibers every night, they had to use grad students. Probably cheaper, but still less efficient. (Note that I was never one of those unlucky “volunteers” but I did hear horror stories.)

Sure, the DESI survey is less than 1% of all the galaxies in the observable volume of the cosmos, but it’s still pretty sizable. So what do you do with a map of a decent chunk of the entire universe?

I’m glad you didn’t ask, because I’m happy to answer. The arrangements of galaxies on very large scales tells us a lot about the universe. And one of the key things used in this new DESI analysis is a feature of the large-scale universe goes by the ungainly but super nerdy name of baryon acoustic oscillations, or BAO for short.

Check this out. Long ago the universe was much smaller, hotter, and denser than it is today. If you’re ever asked what the big bang theory is all about, that’s pretty much it in a nutshell. In fact, billions of years ago, when the universe was only a few hundred thousand years old, it was so hot and dense (for those of you keeping score at home, a million times smaller than its present volume and thousands of degrees hotter) that all the matter was crammed together in the form of an energized plasma. This is the same state of matter as the body of the Sun or a lightning bolt, and it literally filled the universe.

Like any dense material, there were sound waves – waves of pressure that crisscrossed the universe. Many of these sound waves were triggered by a competition between gravity and radiation. Dense clumps of matter would try to collapse under their own gravity, but then those clumps would get hot and the radiation they emitted would push them back out.

This seesawing effect went on and on, back and forth, until the plasma cooled down so much that the light was released. This meant that radiation could no longer play the game, and the back-and-forth sound waves got stuck mid seesaw. Wherever they were, they acted as a source of additional gravitation, a shell of slightly higher density.

In fact we even have pictures of these features, which are the baryon acoustic oscillations (or “super hot sound waves” if you prefer). The light that was emitted when this process stopped still exists today, and we can take pictures of it. It’s called the cosmic microwave background, and a decade ago when a bunch of my friends were plugging away their fiber optic cables, I was a member of the Planck collaboration, which was a satellite to map the microwave background.

These shells of extra matter didn’t just go away. They stuck around, and slowly slowly slowly over billions of years more matter accumulated on those shells than the surrounding regions. Today, we see the imprint of the BAO in the form of shells of matter roughly 800 million light-years in diameter.

The cool part about all this is that the shells are what’s called a standard ruler. We know how big the shells are supposed to be – it’s a relatively straightforward calculation to transport the images we see in the microwave background to their sizes in the present day. And we can compare that expected value to how big they appear on the sky. And how big they appear on the sky depends on cosmology: on the properties, history, and evolution of the universe.

The new finding is that the BAO shells found by DESI are a little off. Their sizes don’t quite fit with our usual picture of cosmology. And they seem to fit better a picture of the universe where dark energy is evolving.

But what the heck is dark energy, and why is it so interesting that it might be evolving?