The Day After

What happened after COBE discovered anisotropies in the cosmic microwave background (CMB) launching the era of modern cosmology? How did we get from that day in April, 1992 to today, when we are armed with a fiducial cosmological model ΛCDM? I am so interested in that question because that very model is being threatened by 4 “tensions” that suggest it may need to be replaced. And I keep thinking: how will that happen? So I am looking backwards to try to remember/understand how ΛCDM got anointed in the first place.

Soon after the COBE result was announced, my sense is that almost everyone agreed with the “CDM” part of the model. There had to be dark matter and it had to be slow moving.

The “Cold” part of CDM was necessary in order for structure to form in the universe given the level of clumpiness seen by COBE. The idea is simple and foundational so worth spending a few minutes spelling it out. The CMB is a picture of the early universe, the universe when it was 380,000 years old. From that picture, we learn that the universe was homogeneous: place a sphere of a given radius anywhere in the universe and you would count almost the exact same number of photons, electrons, and protons. The word almost does the heavy lifting: the inhomogeneities depended a bit on the size of the sphere you chose to lay down, but roughly, if there were 10,000 protons in a given region somewhere, there could be 10,001 or 99,998 somewhere else that size but never 99,000 or 101,000 or even 99,500. Very much the same everywhere. Over the next 13 billion years those regions that had slightly more matter in them accreted more and more matter because of gravity. We can actually calculate in standard gravity how much those inhomogeneities would have grown over 13 billion years. And the answer is: they would have not grown enough to make the universe as clumpy as it is today, with galaxies, stars, planets, and people. Dark matter, especially cold dark matter, fixes that problem: the CMB is a picture of how clumpy photons, electrons, and protons were (since they all interacted with one another) but the dark matter was not involved in those interactions so was free to start clumping earlier. Therefore, the level of clumpiness in the dark matter was much larger than that in the ordinary stuff, and it was able to grow into the majestic structure observed today. After electrons and protons combined to form neutral hydrogen atoms, the atoms — the ordinary matter — fell into the much deeper gravitational potential wells occupied by dark matter, so they too were part of the long history of gravitational collapse leading to the formation of structure in the universe.

Long paragraph but important point: Given the level of inhomogeneities measured by COBE, Cold Dark Matter is needed to explain how we got here .

Independent of that argument, dark matter had to exist because if you counted ordinary matter — hydrogen and other elements — you could not account for all the mass observed in galaxies and beyond. Dark matter simply means something else, something besides the observed hydrogen and traces of other elements. The idea had been around for a while and both astronomers and cosmologists had come to accept it. So, the CDM part of ΛCDM was well accepted almost immediately after the COBE result. But the Λ part, the introduction of a cosmological constant in addition to CDM, was not part of the canon.

Title page from one of the many review articles written in the aftermath of COBE. Everyone was talking about what COBE meant, and many people wrote review articles trying to capture these discussions. This one came out in late 1992, less than a year after COBE.

Something else was needed though, as there were problems (or as we would call them today, tensions) with plain vanilla CDM. These were apparent to anybody who attended one of the many conferences held in the aftermath of the COBE announcement. Three meta-comments about those problems:

1. Everybody was talking about them, and everybody knew that CDM alone did not work.

2. In retrospect, no one was sure which problems were most important and therefore should be the pointers to an updated model.

3. There were several possible variations on standard CDM ready to go once point 2 could be answered.

To give people a sense of those 3 points, I have shown in talks some figures and quotes from a review paper I wrote in 1996. Showing my one review article actually undersells the point, as there were many review articles (see above for one of the first) written by many people very soon after COBE. The Liddle and Lyth paper above is a good example of what we were all saying. It has phrases like: “no model has proved as successful as the standard Cold Dark Matter (CDM) model” and “Nevertheless, pressure has recently been exerted on the model …” and “We end by discussing briefly some variants on the CDM model, such as the incorporation of a hot dark matter component, or the introduction of a cosmological constant term.”

So, a partial answer to my opening question is: the experimental result of COBE gave us a solid foundation for what would emerge as the fiducial cosmological model, CDM. But there was also an understanding that CDM by itself probably would not work. Theorists were ready with souped up versions of CDM that could step in once we got more experimental clues. Two of the most compelling sets of experiments came out in the late 1990’s, early 2000’s. I’ll touch on those next time, but spoiler alert: Those did not fix the fiducial model as ΛCDM, although they did point to the need for dark energy. More next week …