Stokes Parameters

The polarization of light, though invisible to the human eye, is a fundamental property that carries a wealth of information about its origin and journey. From the glare reflecting off a surface to the faint signals from distant stars, understanding polarization is key to unlocking secrets across science and technology. But how can we capture and quantify such an ephemeral characteristic? This article addresses the challenge of describing every possible polarization state, from perfectly ordered to completely random, using a unified mathematical language.

In the chapters that follow, we will first explore the "Principles and Mechanisms" behind the Stokes parameters, a set of four simple, measurable quantities that form a complete description of polarized light. You will learn how they are defined, how they distinguish between fully, partially, and unpolarized light, and how they can be visualized using the elegant geometry of the Poincaré sphere. We will then transition to "Applications and Interdisciplinary Connections," where we will see this powerful formalism in action. This chapter will demonstrate how Stokes parameters are used as an indispensable tool in diverse fields, from designing optical systems and navigating with polarized skylight to characterizing quantum bits and even testing Einstein's theory of General Relativity against the backdrop of the cosmos.

How do we talk about something as ephemeral as the polarization of light? We can't see it directly. We can't hold it in our hands. And yet, the universe is awash in it, from the glare off a puddle to the faint signals from distant pulsars. The challenge is to invent a language, a set of numbers, that can capture every possible nuance of polarization—from the perfectly ordered laser beam to the chaotic jumble of a sunbeam. The genius of the Stokes parameters, named after the 19th-century physicist George Gabriel Stokes, is that they achieve this with four simple, measurable quantities. They don't just describe light; they give us a way to ask it questions and understand its answers.

Imagine you have a mysterious beam of light. To understand its polarization, you can't just look at it. You have to probe it. The simplest probe we have is a ​​polarizer​​, a filter that only lets a specific kind of polarization pass through. The key insight is that by measuring the intensity of the light that gets through different types of polarizers, we can deduce everything there is to know about the beam's original polarization state.

The Stokes parameters, labeled $S_0, S_1, S_2$, and $S_3$, are precisely the results of such an interrogation. They are the answers to four fundamental questions we can ask the light beam:

1. ​​$S_0$: What is your total intensity?​​ This is the most basic question. We measure the total energy flow of the beam, without any polarizers in the way. It’s the overall brightness of the light, the sum of all its parts.

2. ​​$S_1$: Do you have a preference for horizontal versus vertical?​​ To answer this, we measure the intensity of the light that passes through a horizontal polarizer ($I_H$) and the intensity that passes through a vertical polarizer ($I_V$). The parameter $S_1$ is simply the difference: $S_1 = I_H - I_V$. A positive $S_1$ means a horizontal preference, a negative $S_1$ means a vertical preference, and a zero $S_1$ means no preference between the two.

3. ​​$S_2$: Do you have a preference for +45° versus -45°?​​ This is the same idea, but with our polarizers rotated. We measure the intensity through a polarizer oriented at $+45^{\circ}$ ($I_{+45}$) and one at $-45^{\circ}$ (or $135^{\circ}$) ($I_{135}$). The difference gives us our third parameter: $S_2 = I_{+45} - I_{135}$. This captures the preference for diagonal polarization.

4. ​​$S_3$: Do you have a preference for spinning right versus spinning left?​​ Linear polarization isn't the whole story. Light can also be circularly polarized, with its electric field vector spinning like a corkscrew. We ask this final question using circular polarizers, one that lets through right-circularly polarized light ($I_R$) and one for left-circularly polarized light ($I_L$). The difference is the final parameter: $S_3 = I_R - I_L$. A positive $S_3$ indicates a right-handed spin, and a negative $S_3$ a left-handed one.

These four numbers—($S_0, S_1, S_2, S_3$)—form the ​​Stokes vector​​, a complete and unambiguous description of the light's polarization state.

Let's see how this language works for light that is ​​fully polarized​​, where the polarization state is perfectly defined and unchanging.

Suppose we have a beam of light with total intensity $I_0$ that is polarized purely vertically. What would its Stokes vector be?

• The total intensity is $S_0 = I_0$.
• When we measure with a horizontal polarizer, nothing gets through, so $I_H = 0$. With a vertical polarizer, everything gets through, so $I_V = I_0$. This gives $S_1 = I_H - I_V = 0 - I_0 = -I_0$.
• A $+45^{\circ}$ or a $-45^{\circ}$ polarizer is at a $45^{\circ}$ angle to the vertical polarization. By Malus's Law, each will pass half the intensity, so $I_{+45} = I_0 \cos^2(45^\circ) = I_0/2$ and $I_{135} = I_0 \cos^2(45^\circ) = I_0/2$. Therefore, $S_2 = I_{+45} - I_{135} = 0$.
• Linearly polarized light can be thought of as an equal combination of right and left circular motions. So, $I_R = I_L = I_0/2$, which means $S_3 = I_R - I_L = 0$.

So, for vertically polarized light, the Stokes vector is ($I_0, -I_0, 0, 0$). The language works! The vector tells us the total intensity is $I_0$, and the large negative $S_1$ screams "I am vertically polarized!"

What about light linearly polarized at $+45^{\circ}$? Following the same logic, we'd find its Stokes vector to be ($I_0, 0, I_0, 0$). The preference has shifted entirely from the $S_1$ "axis" to the $S_2$ "axis". And for purely left-circularly polarized light with intensity $S_0=4 \text{ W/m}^2$, we find the vector is ($4, 0, 0, -4$). There's no preference for any linear polarization ($S_1=S_2=0$), but a maximum possible preference for left-handed spin.

Notice a beautiful pattern here. For all these cases of fully polarized light, a remarkable identity holds: $S_0^2 = S_1^2 + S_2^2 + S_3^2$ This isn't a coincidence. It's the mathematical definition of purity. It tells us that the total intensity is fully "accounted for" by some combination of polarization preferences. The light has made up its mind completely.

But what about the light from a candle flame, an incandescent bulb, or the sun? It’s a chaotic, jumbled mess. We call this ​​unpolarized light​​. If we perform our four measurements on it, we'll find that $I_H = I_V = I_{+45} = I_{135} = I_R = I_L$. There is no preference whatsoever. Its Stokes vector is simply ($I_0, 0, 0, 0$). Notice that for this case, $S_0^2 > S_1^2 + S_2^2 + S_3^2$. This inequality is the key to understanding the real world.

Most light isn't perfectly polarized or perfectly unpolarized; it's somewhere in between. This is ​​partially polarized light​​. Think of it as a mixture—a crowd containing some people marching in formation and others wandering about randomly. The incredible power of the Stokes formalism is that it handles these messy, real-world situations with stunning elegance. If you combine two light beams incoherently (meaning their underlying wave trains are independent, like two separate light bulbs), the Stokes vector of the combined beam is simply the ​​sum​​ of the individual Stokes vectors.

Let's take an example. Suppose we create a beam where half the intensity comes from unpolarized light and the other half from vertically polarized light. Let the total intensity be $I_0$.

• The unpolarized part has intensity $I_0/2$, and its Stokes vector is $(\frac{1}{2}I_0, 0, 0, 0)$.
• The vertically polarized part also has intensity $I_0/2$, and its Stokes vector, as we saw, is $(\frac{1}{2}I_0, -\frac{1}{2}I_0, 0, 0)$.

To find the Stokes vector of the final beam, we just add them up: $S_\text{total} = \begin{pmatrix} \frac{1}{2}I_0 \\ 0 \\ 0 \\ 0 \end{pmatrix} + \begin{pmatrix} \frac{1}{2}I_0 \\ -\frac{1}{2}I_0 \\ 0 \\ 0 \end{pmatrix} = \begin{pmatrix} I_0 \\ -\frac{1}{2}I_0 \\ 0 \\ 0 \end{pmatrix}$ The result is ($I_0, -I_0/2, 0, 0$). This vector tells us a story. The total intensity is $I_0$. There's a preference for vertical polarization (since $S_1$ is negative), but it's not as strong as it could be. It’s not $-I_0$; it’s only $-I_0/2$. The beam is only partially committed.

This leads us to a crucial concept: the ​​Degree of Polarization (DOP)​​, which tells us exactly how polarized a beam is. It's the ratio of the intensity of the polarized part of the light to the total intensity. By thinking of any beam as an incoherent sum of a fully polarized part and a fully unpolarized part, we can derive a beautiful formula: $\text{DOP} = P = \frac{\sqrt{S_1^2 + S_2^2 + S_3^2}}{S_0}$ For fully polarized light, the numerator equals $S_0$, so $P=1$. For unpolarized light, the numerator is zero, so $P=0$. For our 50/50 mixture, $P = \sqrt{(-I_0/2)^2}/I_0 = 0.5$. It's 50% polarized, just as we constructed it! This principle allows us to predict the properties of arbitrarily complex mixtures of light.

The relationship $S_0^2 \geq S_1^2 + S_2^2 + S_3^2$ should tickle a memory from geometry class. It looks like the formula for a point being on or inside a sphere. This is no accident. We can visualize any polarization state as a point in a 3D space whose axes are $S_1, S_2$, and $S_3$. This is the famous ​​Poincaré sphere​​.

• The radius of the sphere is the total intensity, $S_0$.
• Any state of ​​fully polarized light​​ lies on the surface of the sphere. The "North Pole" ($S_3=S_0$) is right-circular polarization. The "South Pole" ($S_3=-S_0$) is left-circular polarization. Points on the equator ($S_3=0$) represent all the states of linear polarization: horizontal ($S_1=S_0$), vertical ($S_1=-S_0$), +45° ($S_2=S_0$), and so on.
• Any state of ​​partially polarized light​​ lies somewhere inside the sphere. The distance of the point from the origin, divided by the radius $S_0$, is simply the Degree of Polarization, $P$.
• And what about ​​unpolarized light​​? The state of maximum randomness, with no preference at all? It sits right at the center: the origin ($0,0,0$). All the directional tendencies have averaged out to nothing. It is the point of perfect equilibrium, of perfect indecision.

This geometric picture is incredibly powerful. It turns abstract algebra into tangible intuition. Analyzing the effect of a polarizing element can be visualized as simply moving the point representing the light's state to a new location on or inside the sphere.

The beauty of a great physical idea is that it reveals connections between seemingly disparate concepts. The Stokes parameters do this in spades.

Consider what happens if we physically rotate our measurement apparatus (our polarizers for the $S_1$ and $S_2$ measurements) by an angle $\theta$. How do the new parameters, say $S'_1$ and $S'_2$, relate to the old ones? A careful derivation reveals a strange and wonderful result. For instance, the new parameter $S'_2$ is given by: $S'_2 = S_2 \cos(2\theta) - S_1 \sin(2\theta)$ This, along with a similar expression for $S'_1$, is exactly the formula for rotating a point ($S_1, S_2$) in a plane by an angle of $2\theta$! Why twice the angle? Think about a linear polarizer. If you rotate it by 180 degrees, it's indistinguishable from where it started. The physical state repeats every 180 degrees, not 360. The mathematical space of polarization must reflect this, so it "spins" twice as fast as the physical rotation. This shows that the ($S_1, S_2, S_3$) are not just a random collection of numbers; they form a vector in a special kind of abstract space that is deeply connected to the geometry of rotations.

Finally, the Stokes parameters also provide a bridge to the even more fundamental quantum-mechanical description of light. They can be used to construct the ​​coherency matrix​​, a $2 \times 2$ matrix $J$ whose elements describe the statistical correlations between the $x$ and $y$ components of the light's electric field. This matrix is, in a sense, "under the hood" of the Stokes parameters. A fascinating connection emerges when we calculate the determinant of this matrix: $\det(J) = \frac{1}{4} (S_0^2 - S_1^2 - S_2^2 - S_3^2)$ Look at that expression! It's directly related to the Degree of Polarization. If the light is fully polarized, the term in the parenthesis is zero, and the determinant is zero. This indicates that the field components are perfectly correlated. If the light is unpolarized, the determinant is maximized, indicating a complete lack of correlation.

From simple intensity measurements, to a powerful additive algebra, to an elegant geometric sphere, and finally to the deep statistical correlations of the electromagnetic field, the Stokes parameters provide a unified and beautiful framework. They give us the language we need to not just describe, but to truly understand the rich and varied character of light.

Now that we have acquainted ourselves with the formal machinery of the Stokes parameters, you might be tempted to view them as a mere mathematical convenience—a clever bookkeeping system for the messy business of polarization. But to do so would be to miss the forest for the trees! The true power and beauty of this description lie not in its definition, but in its application. The four Stokes parameters are not just numbers; they are the alphabet of a language spoken by light, a language that carries stories from the heart of a quantum computer to the edge of the observable universe. Let us now become fluent in this language and see what tales it has to tell.

Our first stop is the most tangible one: the optics laboratory. Here, we are not passive observers of light; we are its active sculptors. We use components like polarizers and wave plates to twist, filter, and transform light to our will. Suppose you have a beam of light, perhaps partially polarized in some complicated way, and you wish to change its state. How can you predict the outcome? The Stokes parameters, combined with the Mueller matrix formalism we touched upon earlier, provide the exact recipe.

Imagine passing a beam of light through a quarter-wave plate—a crystal that "slows down" light polarized along one axis compared to the perpendicular axis. If you orient this plate at some angle $\theta$, you are essentially mixing the different types of polarization. A beam that was, say, purely linearly polarized might emerge with a component of circular polarization. The Stokes formalism allows us to calculate this transformation with beautiful precision. We can find, for example, that the new value of the parameter $S_2$ (which measures the balance of $+45^{\circ}$ and $-45^{\circ}$ linear polarization) becomes a specific mixture of the old $S_1$, $S_2$, and the circular polarization parameter $S_3$, with the mixing ratios depending precisely on the plate's orientation angle $\theta$. This isn't just an academic exercise; it's the fundamental principle behind creating circularly polarized light for 3D movie projectors or analyzing stressed materials in engineering. It is the practical, work-a-day power of the Stokes description.

Let's step out of the lab and look up. The light that reaches us from the cosmos has been traveling for years, sometimes billions of years. Its polarization state is a fossil record of its origins and its journey. By deciphering the Stokes parameters of this light, we become cosmic detectives.

Consider an electron spiraling in a vast magnetic field in a distant nebula. This is a tiny, natural particle accelerator, and as the electron is forced into this helical dance, it radiates energy. What is the character of this light? It turns out that the polarization of the emitted "cyclotron radiation" tells a detailed story about the electron's motion and our viewing angle. If we happen to be looking straight down the axis of the helix, we see purely circularly polarized light ($S_3 \neq 0$, $S_1=S_2=0$). If we look at the orbit from the side, in its plane, we see purely linear polarization. For any angle in between, we see a mixture—elliptical polarization. The degree of circular polarization follows a beautifully simple geometric law, depending only on the cosine of our viewing angle $\theta$. Thus, by simply measuring the Stokes parameters of light from a nebula, an astronomer can deduce the orientation of the magnetic fields within it, fields that are completely invisible to a telescope that only measures intensity.

This principle of "polarization by process" is not confined to exotic cosmic objects. It happens right above our heads. When unpolarized sunlight enters the atmosphere, it scatters off air molecules. This process, called Rayleigh scattering, is not perfectly symmetric. More light is scattered sideways than forwards or backward. This geometric preference imprints itself on the polarization of the scattered light. If you look at the blue sky at a $90^{\circ}$ angle from the sun, you will find it is strongly linearly polarized. If you have polarized sunglasses, you can see this effect by rotating them and watching the sky darken and lighten. A careful analysis of this scattering reveals that it can even generate a non-zero $S_2$ or $S_3$ component if the incoming light has a particular character, with the final polarization depending sensitively on the scattering angle. This is not just a curiosity; many insects, like bees, have eyes sensitive to polarization and use the pattern of the sky's polarization to navigate.

Returning from the vastness of space to the modern technology hub, we find the Stokes parameters are just as essential. In the burgeoning field of quantum computing, a single photon's polarization can be used to represent a quantum bit, or "qubit". The horizontal and vertical polarization states might represent the classical bits 0 and 1. But a quantum state can also be a superposition of these. How do we describe such a state? The Bloch sphere, a geometric representation of a qubit's state, is mathematically identical to the Poincaré sphere. The three Stokes parameters ($S_1, S_2, S_3$), normalized by the total intensity $S_0$, become the three coordinates of a vector pointing to a location on this sphere, perfectly defining the quantum state. Measuring the Stokes parameters of single photons is how we perform "quantum state tomography"—a complete characterization of our qubit. Of course, the real world is imperfect. A wave plate used in a measurement might have a small error in its construction. The Stokes formalism allows us to account for this precisely, showing how an error in a device leads to a predictable "mixing" of the true Stokes parameters into the ones we measure. This allows us to calibrate our quantum devices with exquisite accuracy.

The world of high-speed communications also relies on these ideas. Light signals carrying our internet data race through millions of miles of optical fibers. At the enormous intensities used in modern telecommunications, the fiber itself responds to the light passing through it. This is the realm of nonlinear optics. One such phenomenon is the optical Kerr effect, where the fiber's refractive index changes in proportion to the light's intensity. What's fascinating is that this effect is sensitive to polarization. The presence of circular polarization, measured by $S_3$, can cause the polarization ellipse itself to rotate as the beam propagates. This is a wonderfully self-referential process: the polarization state dictates its own evolution. A detailed analysis shows that the rate of this rotation is directly proportional to the nonlinear coefficient of the fiber and the Stokes parameter $S_3$. Understanding and controlling such effects is paramount to pushing the boundaries of global communication speed and capacity.

We now arrive at the most profound and mind-stretching application of all, where the simple description of polarized light meets the deepest laws of nature: Einstein's theory of relativity.

First, let's consider a simple question within Special Relativity. If an astronomer in a stationary observatory measures the Stokes parameters of a star's light, and an astronaut flies past in a relativistic spaceship and measures the same light, will they agree on its polarization? The answer, astonishingly, is no. Polarization is not an absolute property; it is relative to the observer's state of motion. The rules for how the electric and magnetic fields transform between moving frames dictate a specific, and rather strange, transformation for the Stokes parameters. A beam that one observer sees as purely linearly polarized ($S_3=0$) might be seen by the moving observer as elliptically polarized ($S'_3 \neq 0$). The parameters get mixed and rescaled by factors involving the Lorentz factor $\gamma$. This shows that the Stokes parameters are not just an arbitrary collection of four numbers. They form a structure that has a definite (if complex) behavior under the Lorentz transformations that lie at the heart of spacetime physics.

The story culminates with General Relativity. One of the greatest discoveries of modern science is the Cosmic Microwave Background (CMB), the faint afterglow of the Big Bang. This ancient light is not uniform; it has tiny temperature fluctuations. In the early universe, this light was constantly scattering off free electrons. Just like the Rayleigh scattering that polarizes our sky, this Thomson scattering in the early universe converted the temperature anisotropy of the radiation field into a polarization anisotropy in the light we see today. The resulting map of the sky's polarization, described point-by-point with Stokes parameters, is one of the richest datasets in all of science, encoding secrets about the birth of the cosmos.

But there is one final, glorious twist. That light has been traveling to us for 13.8 billion years, through a universe whose spacetime fabric is curved and warped by the gravity of galaxies and dark matter. General Relativity predicts that as a polarized light ray traverses a gravitational field, its plane of polarization will be dragged along with the curvature of spacetime itself—a phenomenon called geodetic precession. This means that even if a beam of light starts with a fixed polarization, its Stokes parameters $Q$ and $U$ will evolve and rotate into one another simply because of the gravity it passes through. The rate of this rotation is directly related to the curvature of spacetime along the light's path. By measuring the polarization of distant objects, we can literally map the gravitational warps in the fabric of the universe.

From a tool on a lab bench to a ruler for the cosmos, the Stokes parameters provide a unified and powerful language. They demonstrate a beautiful principle in physics: that a well-chosen mathematical description often reveals connections that were previously hidden, weaving together disparate fields into a single, coherent, and magnificent tapestry.