Why we still don’t know how many stars are in the Milky Way

It’s as simple and straightforward a question about our cosmic home as you can imagine: how many stars are in our Milky Way? While there’s an enormous amount that we know about our Universe with a significant amount of confidence — what types of matter and energy are present within it, how much of each species we have, how fast the Universe is expanding, how much time has elapsed since the hot Big Bang, how many galaxies are within it, etc. — learning some critical information about our own home within the cosmos has been exceedingly difficult. Although we live within the Milky Way, there’s a tremendous amount of information about it that we simply struggle to determine.

Our best picture of our galaxy, for example, requires an extensive amount of guesswork, as the galactic plane of the Milky Way is heavily dust-obscured, making it difficult to understand the structure of the far side of our galaxy. We know about our central bulge and bar, as well as our distance to the galactic center and its supermassive black hole, but we struggle with the shape, density, and thickness of our galaxy’s spiral arms and other properties of our galactic disk.

But perhaps the greatest, and simplest, fact that’s still unknown about the Milky Way is how many stars are present within it. While the most commonly cited figure is around 200 billion stars, some sources still cite a lower value of as little as 100 billion, while other sources prefer a higher value: perhaps as great as 400 billion. How can we know so little about our own galaxy, given our extensive knowledge of the Universe? There are major scientific reasons for our ignorance, and our uncertainty reflects the price we pay for being scientifically honest about what we have yet to figure out about our Universe.

The spiral galaxy UGC 12158, with its arms, bar, and spurs, as well as its low, quiet rate of star formation and hint of a central bulge, may be the single most analogous galaxy for our Milky Way yet discovered. It is neither gravitationally interacting nor merging with any nearby neighbor galaxies, and so the star-formation occurring inside is driven primarily by the density waves occurring within the spiral arms in the galactic disk.

Above, what you see is not the Milky Way, but rather a galaxy that shares many features in common with it: a modest central bar and bulge, two large, winding spiral arms with spurs coming off of them, a gas-rich disk, and a comparable diameter to our own of around 100,000 light-years. It’s not experiencing a major, distortive interaction with any other large neighbors, and its star-formation rate is low and quiescent: ongoing largely due to density waves occurring within its spiral arms. This galaxy, like many that are similar to it, may be an excellent candidate for the closest “twin” to our Milky Way of all the galaxies that are out there.

And yet, we don’t “know” for certain that this is at all reflective of the overall shape of the Milky Way and the large-scale features present within it. It’s sort of like being trapped in a world with no mirrors, cameras, or reflective surfaces, and being asked, “What do your own eyes look like?” It’s a mind-boggling question, because your eyes are the things you use to see: you look out from there. Although you’re able to see what everyone else’s eyes look like, and others can see your eyes, it’s impossible to see your own eyes without some sort of reflection, or some sort of externally acquired image, to inform you.

Behind the dome of a series of European Southern Observatory telescopes, the Milky Way towers in the southern skies, flanked by the Large and Small Magellanic Clouds, at right. Although there are several thousand stars and the plane of the Milky Way all visible to human eyes, there are only four galaxies beyond our own that the typical unaided human eye can detect. We did not know they were located outside of the Milky Way until the 1920s: after Einstein’s general relativity had already superseded Newtonian gravity.

Fortunately, we can take all sorts of measurements that do provide us with a great wealth of information about the galaxy that we do inhabit. We can examine the plane of the Milky Way in a wide variety of wavelengths of light. Instead of being restricted to the optical part of the spectrum, where the gas, dust, and clouds of neutral matter within the galactic plane block and obscure the light sources (such as stars) that are located behind that material, we can look in a wide variety of different wavelengths of light, each one revealing a unique set of details about the Milky Way and what’s within it.

• When we look at the shortest wavelengths, in the range of gamma-rays and X-rays, we see active black holes, neutron stars, and extremely hot plasmas, revealing everything from pulsars to flares at the galactic center.
• When we look at optical wavelengths, we see stars in the foreground of this gas and dust, but only silhouettes of the objects behind them, as visible light is largely blocked by neutral matter.
• And when we go to longer wavelengths than the human eye can see — at infrared, microwave, or even radio wavelengths — we can see the light-and-heat emitting sources, all the way down to temperatures that are only a few degrees above absolute zero, where these long-wavelength signals are transparent to the gas and dust that block visible light.

This animation shows the Bok globule Barnard 68 in a variety of visible and infrared wavelengths. As the longer wavelengths reveal, this is not a hole in the Universe but simply a dusty cloud of gas, where the longer (redder) wavelengths of light penetrate and pass through the dust. As dust clouds form and dissipate, the dust density can be revealed by examining the light blocked and transmitted by fixed, background objects.

Most of our astronomical efforts, with telescopes, focus on a narrow region of the sky, as shown above for the Bok globule Barnard 68. The above animation shows the same region of sky — the same stars, the same gas, and the same dust features — viewed at a variety of wavelengths of light: from visible light at 440 nanometers (0.44 microns) all the way up to near-infrared wavelengths of 2160 nanometers (2.16 microns). For comparison, the visible light range spans from 400-700 nanometers and infrared wavelengths are longer-wavelength than that. As you can see, the gas and dust are extremely efficient at blocking optical (visible) light, but do worse and worse at blocking light the longer the wavelength we choose.

This is because the light-blocking material that the galaxy is filled with is largely composed of grains of dust of a finite size: a size comparable to the wavelength of light. In general, dust is very efficient at blocking light of wavelengths that are shorter than the size of the dust grains, and progressively less efficient at blocking light of longer wavelengths. (This is why dust-obscured regions appear redder, and dust-free regions appear bluer.) If we apply this technique — of looking at the whole sky, including the plane of the Milky Way, at longer, infrared wavelengths of light — we can do an outstanding job of “seeing through” the light-blocking gas and dust, and thereby reveal the stars that are located both within and behind those dust clouds, dust lanes, and any other shaped features that dust can produce.

By viewing the Milky Way in infrared wavelengths of light, we can see through large amounts of the galactic dust and view the distribution of stars and star-forming regions behind them. As revealed by the 2 micron all-sky survey (2MASS), the densest collections of galactic dust can be seen tracing out our spiral arms, but the center of the plane of the Milky Way is where the dust is densest. Infrared and visible light views both showcase this, but in vastly different ways.

By peering through the plane of our Milky Way, including:

• into the galactic center and its surrounding regions,
• through the gas and dust and down, directly, the spiral arms within the galactic plane,
• away from the galactic center and toward the outskirts of the Milky Way,
• and even clear through to the opposite side of the galaxy from the side the Sun is located within,

we can learn an enormous amount of information about the stellar extent of the Milky Way. This includes information about how large the stellar extent of the Milky Way is, whether it’s a straight disk or whether there’s a warp to it, what the stellar density is at all radii away from the galactic center, on average, as well as how the types and ages of stars present evolve as a function of distance from the galactic center.

By the mid-2010s, this led us to a profound conclusion about the total amount of mass that’s present in the form of stars within the Milky Way: about 60 billion solar masses worth of stars, with an uncertainty of around ~20% at that time. That’s an incredibly precise figure, as we had gathered enough information about the light in our Milky Way, even the light that gets blocked at optical wavelengths, to understand how much of it there was, and what the various numbers and types of stars were needed to produce them.

The (modern) Morgan–Keenan spectral classification system, with the temperature range of each star class shown above it, in kelvin. In terms of size, the smallest M-class stars are still about 12% the diameter of the Sun, but the largest main sequence stars can be dozens of times the Sun’s size, with evolved red supergiants (not shown) reaching hundreds or even 1000+ times the size of the Sun. A star’s (main sequence) lifetime, color, temperature, and luminosity are all primarily determined by a single property: mass.

And yet, that small uncertainty in stellar mass still translates into a tremendously large uncertainty for the total number of stars present in the Milky Way: an uncertainty where pretty much all “guesstimates” between 100 billion and 400 billion stars were still viable.

How is this the case?

Because even a very precise, low-uncertainty figure for overall stellar mass (and stellar light) translates into a very large uncertainty for the population of the most common type of star: the faint, cool-temperature, low-mass, M-class red dwarf stars. There are the most common types of stars of all, and yet they emit far less light, even all combined, than the bigger, brighter, and rarer high-mass stars they’re born alongside.

Think about this non-intuitive fact: if you were to take a star like our Sun and either double or halve its mass, how much brighter or fainter would you expect that new star, of either 2 or 0.5 solar masses, to be? Do you think the twice-as-massive star will be twice as bright? Do you think the half-as-massive star will be only half as bright? Unsurprisingly, in order to know for sure, we have to get the relevant data directly.

• Consider Sirius A: a star just 8.6 light-years away, with a mass that’s 2.063 solar masses. It’s a whopping 25 times as bright as our Sun.
• Then consider Alpha Centauri B, just 4.3 light-years away also known as Toliman. It is indeed a star that’s half as luminous as our Sun, but it still has, impressively, 91% of the Sun’s mass.
• If we look at a star like Lacaille 9352, 10.7 light-years away, which is half the Sun’s mass, you might be surprised to learn that it’s only ~4% as intrinsically bright as our Sun is.

This photo showcases Proxima Centauri: the closest star to our own Sun at present. Although it’s only 4.24 light-years away, Proxima Centauri is not even close to visible to the naked eye, as it’s intrinsically nearly 1000 times fainter than the Sun.

So what happens, then, when we start thinking about the lowest-mass stars of all? What happens when we start considering stars like Proxima Centauri, shown above, which is actually the closest star to our Sun across the entire sky?

First off, despite being the closest star to the Sun, it’s not visible to the naked eye. In fact, it’s so faint that it wasn’t even discovered until 1915: just 110 years ago. At about 12% the mass of our Sun, it’s only about ⅐^th the radius of the Sun, but much, much fainter than that. Intrinsically, Proxima Centauri is only 0.15% as luminous as the Sun, with nearly all of its energy being emitted in the infrared part of the spectrum. If we restricted ourselves to looking only in the visible light portion of the spectrum, we’d find that Proxima Centauri is only 0.005% as bright as the Sun, implying that it would take 20,000 stars similar to Proxima Centauri to appear as optically bright as the Sun.

This is where the greatest source of uncertainty lies: in the fact that we cannot measure the number of low-mass stars in the Universe, or in our galaxy, directly. Even the world’s most advanced, dedicated scientific collaboration to attempt to detect and find these low-mass stars, RECONS (the REsearch Consortium On Nearby Stars), has only been able to survey a tiny volume of space close to the Sun in its quest for the lowest-mass stars of all. Of the 270 star systems within 10 parsecs (33 light-years) of Earth that contain a “true star” within them, 82% of those systems are dominated by a red dwarf (M-class) star, with no brighter or more massive stellar member present at all.

This graphic compares a Sun-like star with a red dwarf, a typical brown dwarf, an ultra-cool brown dwarf, and a planet like Jupiter. Only about 5% of all stars are like the Sun or more massive; K-type stars represent 15% of all stars, while red dwarfs represent 75-80% (or more) of all stars. Brown dwarfs, although they are failed stars, may be just as common as red dwarfs are, but are cooler and lower in mass.

In other words:

• we can make a good estimate for the total amount of stellar mass, or mass in the form of stars, in the Milky Way, simply by measuring the total amount of starlight,
• which we can do directly from surveys, including all-sky surveys, including in infrared wavelengths of light,
• and this tells us to a great degree of accuracy the number and abundance of Sun-like stars, more massive stars than the Sun, and cooler stars than the Sun down to about 40-50% of the Sun’s mass,
• but that at the low-mass end of stars, including objects between 7.5-40% the mass of the Sun, tremendous uncertainties in the number and abundance of stars remain.

It’s the lowest-mass end of the stellar spectrum that houses the greatest uncertainty as far as their numbers go, because they contribute such a tiny, practically negligible amount of starlight (and mass) to the total stellar budget of the Milky Way.

However, all hope is not lost. One of the most powerful and compelling science missions of the last decade, in terms of how much it can teach us about our local Universe and our place within it, is the ESA’s Gaia mission. Located in space, Gaia observed the 3D positions and properties of nearly 2 billion stars within the Milky Way, determining their distance from us to better precision than any survey or catalogue before it. Gaia’s map of the stars within the Milky Way — the best one ever — has resulted in a number of new discoveries: the Milky Way’s warped disk, the full extent of its stellar disk, and superior distance determinations to a variety of features.

Gaia’s all-sky view of our Milky Way Galaxy and neighboring galaxies. The maps show the total brightness and color of stars (top), the total density of stars (middle), and the interstellar dust that fills the galaxy (bottom). Note how, on average, there are approximately ~10 million stars in each square degree, but that some regions, like the galactic plane or the galactic center, have stellar densities well above the overall average.

One of the big advances that came along with Gaia, although it wasn’t necessarily well-publicized, was the best-ever estimate of the stellar mass of the entire Milky Way itself: about 50.4 billion solar masses worth of stars, with an uncertainty of approximately ~9-10% on that figure. This is consistent with, but a more precise and more refined version of, all previous estimates of the stellar mass of the Milky Way. This represents the best estimate of our galaxy’s stellar mass of all-time: the total amount of mass in the form of stars within the Milky Way. (Gaia also measures the Milky Way’s total mass, including dark matter.)

Interestingly, this produces a total amount of light that corresponds to about 150-180 billion Suns. How is this possible? Because the more massive, more luminous stars than the Sun dominate the light output of the stars within the Milky Way. In fact, there are several stars located within our Milky Way, including:

are all only a few dozen to perhaps a couple of hundred times as massive as the Sun, but are millions of times as luminous as the Sun is. The more massive, bluer stars in the Milky Way dominate its light output, despite being so few in number, while the more numerous, redder, but less massive stars dominate the total “number of stars” within the Milky Way, but produce very little of the Milky Way’s total light.

Various estimates for the Milky Way’s stellar mass, over time, with the latest (2020) data informed by the ESA’s Gaia mission. With a stellar mass of ~50 billion Suns, but hundreds of billions of stars inside the Milky Way, it’s the red dwarf stars that create the largest uncertainties in the total number of stars within our galaxy.

And that’s where we are today. We understand very well the total amount of light produced by the stars within the Milky Way: produced largely by the small percentage of stars that are greater in mass than the Sun, and augmented by Sun-like and slightly less massive stars. This understanding enables us — particularly with data acquired from the ESA’s Gaia mission about the stars in the Milky Way, directly — to draw compelling conclusions about the total amount of stellar mass, or mass in the form of stars, present in the Milky Way. Our best estimate for that is ~50 billion solar masses worth of stars, with an uncertainty of just ~10%.

However, our lack of understanding about the lowest-mass end of the stellar spectrum, and how many red dwarf stars there are of between 7.5-40% the mass of the Sun, results in tremendous uncertainties down there. With the latest data, the total number of stars in the Milky Way could reasonably be as low as ~200 billion, but not as low as the ~100 billion estimate that so many astronomers used in the late 20th century. However, it still could be as high as ~400 billion; the data remains consistent with that larger number as well.

The fact that these low-mass stars are so numerous and abundant, but also so faint and difficult to directly detect, is the largest source of uncertainty surrounding the number of stars within the Milky Way. Until we’ve overcome that limitation, we’ll simply have to reckon with the fact that there could be twice as many stars as the ~200 billion figure astronomers commonly use as an estimate for the Milky Way, and only by acquiring more and better data will we be able to know for certain.