Tully-Fisher Relation

In the vast expanse of the cosmos, astronomers seek simple rules to make sense of staggering complexity. The Tully-Fisher relation is one such rule—a surprisingly tight and powerful correlation between how fast a spiral galaxy spins and how bright it intrinsically is. But why should this connection exist? Is it a mere coincidence, or does it point to a deeper physical truth about how galaxies are built and evolve? This article delves into this cosmic mystery, addressing the fundamental physics that underpins this empirical law.

The following chapters will guide you through this fascinating topic. First, in "Principles and Mechanisms," we will explore the physical foundations of the relation, building it from simple gravitational principles and examining the crucial roles played by dark matter, galaxy stability, and even alternative theories of gravity like MOND. Then, in "Applications and Interdisciplinary Connections," we will see how this relation is wielded as a master key in astronomy, used not only as a cosmic yardstick to measure the universe but also as a probe to weigh invisible dark matter halos and test the very laws of physics across cosmic time. Let's begin by unraveling the physics behind this magical cosmic yardstick.

Now that we've been introduced to the Tully-Fisher relation as a magical cosmic yardstick, you might be wondering, "Why on earth should it work?" Why should the speed at which a galaxy spins have anything to do with how bright it is? Is this just a lucky coincidence, a mere empirical curiosity? Or is it telling us something profound about the way galaxies are built? As we peel back the layers, we'll find that this simple-looking rule is a thread that connects gravity, the mysterious nature of dark matter, the life cycle of stars, and even challenges our understanding of the fundamental laws of physics. Let's embark on a journey, much like a physicist would, starting with the simplest possible model to see if we can "invent" the Tully-Fisher relation ourselves.

Let’s try a thought experiment. Imagine a spiral galaxy is just a simple, spinning plate of stars. What holds it together? Gravity. What keeps it from collapsing into a single point? Its rotation. For a star at some characteristic distance $R$ from the center, moving at the flat rotation speed $v_{\text{flat}}$, the inward pull of gravity must exactly balance the outward centrifugal force. This is the same physics that keeps a satellite in orbit around the Earth. The gravitational pull comes from the total mass $M$ inside that radius. This gives us a familiar equation:

$v_{\text{flat}}^2 = \frac{G M}{R}$

This equation connects kinematics ($v_{\text{flat}}$) to mass ($M$). But the Tully-Fisher relation connects velocity to luminosity ($L$). So, we need a bridge between mass and light. Let's make a simple assumption: what if the amount of light a galaxy produces is just proportional to its mass? This seems reasonable; more stuff should mean more stars, and thus more light. We can write this as $M = \Upsilon L$, where $\Upsilon$ (the Greek letter Upsilon) is the ​​mass-to-light ratio​​.

Now we have $v_{\text{flat}}^2 \propto L/R$. We’re getting closer, but we have this pesky radius $R$ in the way. How can we get rid of it? Let's make one more bold, simplifying assumption, inspired by an observational fact sometimes called Freeman's Law. Many spiral galaxies seem to have a remarkably similar brightness per unit area, or ​​surface brightness​​. Let's imagine an idealized world where this is perfectly true—that all spiral galaxies are painted with the same intrinsic "luminosity density," $\Sigma_L$. The total luminosity of our disk galaxy out to radius $R$ would then simply be its area times this constant surface brightness: $L = \pi R^2 \Sigma_L$.

Now we have a way to relate $R$ and $L$. If we rearrange this, we find $R \propto \sqrt{L}$. Let’s plug this back into our force-balance equation:

$v_{\text{flat}}^2 \propto \frac{L}{R} \propto \frac{L}{\sqrt{L}} = \sqrt{L}$

If we square both sides, we find $v_{\text{flat}}^4 \propto L$. And there it is! Under a few simple, physically-motivated (though idealized) assumptions, we have derived the Tully-Fisher relation: ​​Luminosity is proportional to the fourth power of the rotation velocity​​. It’s not magic; it’s a consequence of the interplay between Newtonian gravity, the geometry of a disk, and the relationship between mass and light.

Our simple model is satisfying, but it hides a skeleton in the closet. When we actually go out and measure the rotation speeds of stars in galaxies, we find something astonishing. The speeds don't drop off at large radii as our simple model would predict if only the visible stars and gas were providing the gravity. They stay remarkably constant, or "flat," far out into the galactic halo. This is the primary evidence for ​​dark matter​​, an invisible substance that provides the extra gravitational glue. In fact, in the outer parts of a galaxy, the dynamics are almost completely dominated by this unseen halo.

Does this ruin our neat derivation? Let's see. We can build a new model where the stars are just a sprinkling of light within a vast, dominant halo of dark matter. A very common and simple model for such a halo is the ​​Singular Isothermal Sphere (SIS)​​, which has a density that falls off as $1/R^2$. This particular density profile has a wonderful property: it naturally produces a perfectly flat rotation curve, where the velocity $v_c$ is constant everywhere. This is exactly what we need!

So now, the velocity $v_c$ is set by the dark matter halo, not the stars. But the luminosity $L$ still comes from the stars. How can there be a connection? Well, it's reasonable to assume that a bigger, more massive dark matter halo would have been able to attract more gas in the early universe and thus form a more massive and luminous galaxy. Let’s formalize this by assuming that the mass of the stars, $M_*$, is proportional to the mass of the dark matter, $M_{DM}$, within the same radius. If we again assume that galaxies have a constant stellar surface density, we can work through the math. The result is uncanny: we once again find that $L \propto v_c^4$.

This is a crucial insight. The Tully-Fisher relation holds whether the galaxy's mass is mostly in its stars (as in our first model) or mostly in a dark matter halo. This robustness tells us the relation is not just a fluke of one particular model, but reflects a deep regularity in how galaxies form and are structured within their dark matter halos.

So far, we've used luminosity ($L$) as a proxy for the "stuff" that makes up a galaxy. But this can be a fickle measure. The light we see is dominated by massive, bright, young stars. A galaxy that has recently undergone a burst of star formation will be much brighter than a galaxy of the same mass that has been quietly sitting for billions of years. This variation is captured in the mass-to-light ratio, $\Upsilon$. Different star formation histories lead to different values of $\Upsilon$, which in turn introduces significant "scatter" into the Tully-Fisher relation—making it less precise.

This suggests that luminosity isn't the most fundamental quantity. A physicist always seeks the more fundamental, conserved quantity: mass. The "normal" matter that stars, planets, and we are made of is collectively called ​​baryonic matter​​. In a galaxy, this is overwhelmingly composed of stars ($M_s$) and cold gas ($M_g$). Perhaps the true fundamental relationship is not between light and velocity, but between the total baryonic mass ($M_b = M_s + M_g$) and velocity.

This leads us to the ​​Baryonic Tully-Fisher Relation (BTFR)​​:

$M_b \propto v_{\text{flat}}^4$

This relation is indeed observed to be much "tighter," with less scatter, than the original version based on luminosity. It's a more direct link between the total payload of normal matter in a galaxy and the depth of the gravitational well (probed by $v_{\text{flat}}$) it sits in.

The importance of using baryonic mass is not just academic. Imagine you have calibrated a distance-finding tool using the stellar mass Tully-Fisher relation on a sample of "average" galaxies. Now, you observe a new galaxy that happens to be very rich in gas. Because much of its baryonic mass is in the form of non-luminous gas, its stellar mass (and luminosity) will be deceptively low for its rotation speed. Applying your old calibration, you would incorrectly infer that the galaxy is closer than it really is. Using the full baryonic mass avoids this systematic error.

We've established that the BTFR seems to be the fundamental law. But we can still ask why. Why this particular relationship? Why the fourth power? An even more profound insight comes from thinking about the stability of a galactic disk. A spinning disk of stars and gas is a delicate balancing act. If it spins too slowly for its own gravity, it will collapse into clumps and bars. If the random motions of its stars (its "temperature") are too low, the same thing will happen. The disk must be "stable" to maintain its grand spiral structure over billions of years.

The physicist A. Toomre developed a criterion for this, encapsulated in a parameter $Q$. For a disk to be stable, $Q$ must be above a certain critical value. It turns out that many galactic disks seem to hover right around a state of ​​marginal stability​​.

Let's imagine a family of galaxies that all have the same central surface mass density and are all marginally stable. If we combine the physics of Toomre stability with the dynamics of a flat rotation curve, we can once again derive the relationship between the total baryonic mass and the flat rotation velocity. Remarkably, out of this physically rich and complex picture of disk stability, the same simple law emerges: $M_b \propto v_{\text{flat}}^4$. This is a beautiful piece of physics. It suggests that the Tully-Fisher relation is not just about how galaxies form, but is also a direct consequence of the physics that keeps them in their observed, stable state.

For decades, the standard explanation for flat rotation curves, and thus the foundation of the Tully-Fisher relation, has been dark matter. But what if we've got it wrong? What if the problem isn't a missing substance, but a misunderstanding of gravity itself? This is the core idea behind ​​Modified Newtonian Dynamics (MOND)​​, proposed by Mordehai Milgrom.

MOND suggests that Newton's law of gravity, or his law of inertia, is perfectly correct in the high-acceleration environments we're used to (like our solar system), but it changes in the realm of incredibly tiny accelerations found in the outskirts of galaxies. In this "deep-MOND regime," the theory posits that the true acceleration $a$ experienced by a body is related to the standard Newtonian gravitational acceleration $g_N$ by a simple law: $a^2 = g_N a_0$, where $a_0$ is a new fundamental constant of nature.

Let's see what this means for a star in a circular orbit on the edge of a galaxy. Its true acceleration is the centripetal acceleration, $a = v_{\text{flat}}^2/r$. The Newtonian acceleration it would feel is from the total baryonic mass of the galaxy, $g_N = GM_b/r^2$. Let’s plug these into the MOND relation:

$\left(\frac{v_{\text{flat}}^2}{r}\right)^2 = \left(\frac{G M_b}{r^2}\right) a_0$

$\frac{v_{\text{flat}}^4}{r^2} = \frac{G M_b a_0}{r^2}$

The $r^2$ terms on both sides magically cancel! We are left with:

$M_b = \frac{1}{G a_0} v_{\text{flat}}^4$

This is astounding. MOND doesn't just accommodate the Baryonic Tully-Fisher Relation; it predicts it from first principles. It predicts the exponent must be exactly 4, and even gives a theoretical value for the proportionality constant in terms of $G$ and the new constant $a_0$. Furthermore, this framework naturally explains why there's a tight relation for pressure-supported systems like elliptical galaxies too, providing a unified explanation for both types of galaxies. The fact that the Tully-Fisher relation falls out so naturally from MOND is considered by its proponents to be one of the theory's greatest successes. It forces us to confront the possibility that the "ghost in the machine" might not be dark matter, but our cherished law of gravity.

Our theoretical journey has shown that a $v^4$ law for galaxies seems to be woven into the fabric of the cosmos, whether through dark matter dynamics, stability physics, or modified gravity. However, when we look at real data, we don't see a perfect, infinitely thin line. We see a narrow cloud of points scattered around the central trend. This "scatter" isn't just measurement error; it is itself a source of physical information.

What causes it? As we've discussed, variations in the mass-to-light ratio are one source. Another source could be a "velocity bias"—the possibility that the visible galaxy doesn't spin at exactly the characteristic speed of its host dark matter halo. Furthermore, galaxies are not all simple, pure disks. Many have a central ​​bulge​​. A galaxy with a large, massive bulge will have a different mass distribution than a pure disk galaxy of the same total baryonic mass, causing it to lie systematically offset from the main relation for pure disks.

Finally, even the act of measuring the velocity $v$ is fraught with physical corrections. The Doppler-broadened line width we observe from a galaxy's gas has to be corrected for the galaxy's inclination to our line of sight, as well as for the random pressure and turbulence in the gas, a phenomenon known as asymmetric drift.

Far from being a problem, this messiness is an opportunity. By studying the scatter and the systematic deviations from the ideal Tully-Fisher relation, we can learn about the diversity of galaxy formation histories, the structural makeup of galaxies, and the intricate relationship between the visible and dark components of the universe. The Tully-Fisher relation, in all its elegant simplicity and messy reality, remains one of our most powerful tools for understanding the grand machinery of the cosmos.

After our journey through the principles and mechanisms of the Tully-Fisher relation, you might be left with a feeling of "So what?". It is a beautiful empirical fact that the spin of a galaxy is tied to its brightness. But what is it good for? It turns out this simple correlation is not just a curiosity; it is a master key, capable of unlocking some of the deepest secrets of the cosmos. Its applications stretch from our own cosmic backyard to the very edge of fundamental physics and the dawn of time. Let us now embark on a tour of these applications, to see how this one relation weaves together seemingly disparate threads of the cosmic tapestry.

The most immediate and famous use of the Tully-Fisher relation is as a cosmic yardstick. If you know a galaxy's intrinsic luminosity—its true wattage—and you measure its apparent brightness in your telescope, the inverse-square law tells you its distance. The Tully-Fisher relation is a magical machine for determining that intrinsic luminosity. You simply point your spectrograph at a spiral galaxy, measure the Doppler shift across its disk to find its maximum rotation speed, $v_{max}$, and the relation tells you its luminosity, $L$.

But how do we build and trust such a ruler? This is where the beautiful, self-reinforcing structure of astronomical measurement comes into play. We must first calibrate our ruler. We do this by finding nearby galaxies whose distances we already know through other, more direct means. For instance, we can find galaxies close enough to resolve individual Cepheid variable stars—pulsating stars whose period of pulsation is rigidly tied to their luminosity. By measuring the periods of these stars, we find their luminosity, calculate the galaxy's distance, and then measure that same galaxy's rotation speed. Do this for a few dozen galaxies, and you have calibrated your Tully-Fisher ruler. Of course, a true scientist is never satisfied with just a number; they must know how well they know it. Every measurement, whether of a Cepheid's brightness or a galaxy's rotation, has an uncertainty. A careful analysis allows us to combine all these uncertainties to determine the precision of the Tully-Fisher relation itself, ensuring our cosmic ruler doesn't wobble.

This process is a wonderful illustration of the "Cosmic Distance Ladder," where each rung allows us to climb to the next. We can even use one powerful distance indicator to calibrate another. For galaxies a bit further out, we might find one that has recently hosted a Type Ia supernova—a stellar explosion so uniform that it can be used as a "standard candle." The observed brightness of the supernova gives us the distance to its host galaxy, which we can then add as a new calibration point to our Tully-Fisher relation, extending our reach even deeper into the cosmos.

However, no tool is without its limitations, and understanding them is as important as understanding its use. To get the true rotation speed, $v_{max}$, we must see the galaxy's disk at least somewhat edge-on. A galaxy seen perfectly face-on is like a spinning coin viewed from directly above; all the motion is transverse, and we can't measure any rotation speed from the Doppler effect. The speed we measure, $v_{obs}$, is always $v_{obs} = v_{max} \sin(i)$, where $i$ is the inclination angle. Getting that angle wrong—mistaking a slightly tilted galaxy for one that is nearly edge-on, for instance—can lead to a significant error in the inferred luminosity, and thus a large error in the distance. The art of the astronomer lies in skillfully accounting for these practical difficulties.

For a long time, the Tully-Fisher relation was primarily a tool for cosmography—mapping the universe. But a deeper truth was waiting to be discovered. The relation becomes far more profound when we consider not just the light from a galaxy, but its total mass in ordinary, or baryonic, matter (stars and gas). This is the Baryonic Tully-Fisher Relation (BTFR), which connects the total baryonic mass, $M_b$, to the rotation speed, $v_{max}$. This version is more fundamental because it deals with mass, the ultimate source of gravity, rather than light, which can be affected by messy details like dust and recent bursts of star formation.

And here is where the story takes a dramatic turn towards the invisible. The rotation speed of a galaxy is a measure of the total gravitational pull holding it together. If baryonic matter were all there was, the BTFR would simply be a statement about self-gravitating disks. But we know this is not the case. The flat rotation curves of galaxies tell us they are embedded in vast, unseen halos of dark matter.

This is the genius of the BTFR: it gives us a way to weigh the darkness. Cosmological models predict that the universe was cooked with a very specific recipe: for every one part of baryonic matter, there are about five parts of dark matter. This universal baryon fraction, $f_b = \Omega_b / \Omega_m$, should be reflected in the composition of large structures like galaxies. If we assume a galaxy is a fair sample of this cosmic mix, we can perform a beautiful piece of cosmic accounting. We measure a galaxy's rotation speed, $v_{max}$. The BTFR tells us its baryonic mass, $M_b$. We then say: "If this galaxy follows the universal recipe, its total mass must be $M_{total} = M_b / f_b$." The mass of the dark matter halo is simply what's left over: $M_{DM} = M_{total} - M_b$. Suddenly, a simple scaling relation for visible matter has become a tool for probing the dominant, invisible component of the cosmos.

The Tully-Fisher relation does not live in isolation. It is a node in a vast network of physical laws and empirical relations that describe how galaxies work. Its connections to these other laws reveal a deep unity in the processes that govern galaxy formation and evolution.

Consider, for example, the great divide between spiral galaxies—cool, rotating disks of gas and young stars—and elliptical galaxies—"hot," puffy spheroids of old stars. They seem like completely different species. Spirals follow the Tully-Fisher relation ($L \propto v_{max}^4$), while ellipticals follow the Faber-Jackson relation, which connects their luminosity to the random velocities of their stars, called velocity dispersion ($L \propto \sigma_v^4$). But perhaps they are more like cousins. In many spiral galaxies, the central bulge itself looks like a small elliptical galaxy. Astronomers have found that the velocity dispersion of the bulge and the rotation speed of the disk are often tightly linked. This suggests that a galaxy can live on both relations at once! The same underlying physics of gravity and mass might be manifesting in two different ways—ordered rotation in the disk, and random motion in the bulge—unifying the dynamics of these different structures under a single framework.

This drive for unification also leads us to ask: where does the Tully-Fisher relation come from? Is it just a brute fact, or can we derive it from more fundamental pieces? Let's try to build it from the ground up. The light from a star-forming galaxy is powered by, well, star formation. A significant portion of this energy heats up dust, which glows in the far-infrared (FIR). But star formation, especially of massive stars, also leads to supernovae. These explosions create cosmic rays that spiral in the galaxy's magnetic fields, producing radio waves. It's a remarkable fact that for galaxies, the FIR and radio luminosities are almost perfectly proportional to each other. So, radio luminosity tracks star formation. Now, what determines the star formation rate? The Kennicutt-Schmidt law tells us it depends on the local density of gas. If we take a simple model for how gas is distributed in a galaxy (an exponential disk) and combine it with the laws of galaxy dynamics that relate a disk's mass and size to its rotation speed, a startling prediction emerges. The pieces all click together to predict a relation of the form $L_{radio} \propto v_{max}^4$. We have derived the Tully-Fisher relation from the physics of star formation! This is a stunning example of the unity of physics, connecting the mechanics of a spinning galaxy to the intricate processes of stellar birth and death within it.

The relation also serves as a bridge between the properties of individual galaxies and the statistical description of the entire galaxy population. Astronomers conduct cosmic censuses, like the "luminosity function," which counts how many galaxies exist at each level of brightness, and the "velocity function," which counts how many galaxies spin at each speed. These are two of the most fundamental measurements of our universe's structure. The Tully-Fisher relation is the conversion factor between them. If you know the velocity function and the Tully-Fisher relation, you can precisely predict the shape of the luminosity function, and vice versa. It is the Rosetta Stone that translates the language of dynamics into the language of luminosity.

We now arrive at the most profound applications of the Tully-Fisher relation—where it ceases to be just a tool of astrophysics and becomes an arbiter in debates about fundamental physics and the history of the universe itself.

The standard model of cosmology, with its vast halos of dark matter, can reproduce the Tully-Fisher relation, but it is a complex, emergent phenomenon involving the messy physics of gas cooling and star formation. It is a curious fact that an alternative theory of gravity called Modified Newtonian Dynamics (MOND)—which proposes that gravity itself behaves differently at the low accelerations found in galaxies—predicts the Baryonic Tully-Fisher relation ($M_b \propto v_{max}^4$) as one of its most basic consequences. The observed relation is so tight and simple that it presents a persistent challenge to the standard model. The Tully-Fisher relation has thus become a key battleground. The two theories make subtly different predictions, and one could imagine a future observation that produces a systematic error in our distance measurements if we assume the wrong underlying theory. By measuring distances and masses with extreme precision, astronomers hope to use the Tully-Fisher relation to test the law of gravity itself.

Furthermore, the universe is not static; it evolves. By looking at galaxies billions of light-years away, we are looking back in time. Did the Tully-Fisher relation exist in the ancient universe? Did it have the same form? Our theories of how dark matter halos grow and become more concentrated over cosmic time predict that for a galaxy of a fixed rotation speed, its mass (and thus luminosity) should have been different in the past. This means the zero-point of the Tully-Fisher relation should evolve with redshift. Observing this evolution is like watching galaxies get assembled over billions of years, providing a crucial test of our models of cosmic structure formation.

Finally, in a beautiful twist, this relation, which we use to measure distances within a given cosmology, can also be used to test the assumptions of that cosmology. Imagine you set out to measure the slope of the Tully-Fisher relation using a sample of galaxies spanning a range of redshifts. The distances you calculate for those galaxies depend on the cosmological model you assume—specifically, its expansion history, which is governed by parameters like the matter density, $\Omega_{m,0}$. If you assume the wrong cosmology (say, a universe with only matter and no dark energy), the distances you calculate will be systematically wrong, and you will, in turn, measure an incorrect slope for the Tully-Fisher relation. This means that a precise measurement of how the apparent slope of the relation changes with distance can actually be used to constrain the parameters of the universe as a whole!

From a humble yardstick to a probe of galaxy evolution, dark matter, and the very laws of gravity, the Tully-Fisher relation is a testament to the power of a simple observation. It reminds us that the universe is a deeply interconnected place, where the way a galaxy spins is tied to the birth of stars within it, the invisible matter that surrounds it, and the cosmic history that shaped it. It is a simple dance of light and motion that echoes the deepest harmonies of the cosmos.