Why is mathematics important for understanding rainbows? A rainbow looks like a coloured arc painted onto distant rain, yet no coloured object occupies that location. Sunlight enters countless nearly spherical droplets, bends, reflects inside and leaves at angles that concentrate light toward an observer. Geometry and refraction explain the arc; calculus explains why some angles are brighter; wavelength-dependent refractive index explains the colour order.
Rainbows are an especially joyful example of mathematics in everyday life because a familiar sight becomes more beautiful when its structure is understood. The equations do not replace wonder. They reveal why the Sun must be behind the observer, why every observer has a personal bow, why red lies outside violet in the primary bow and why a secondary bow reverses the order.
The simple ray model is powerful but not complete. Real sunlight has a spectrum, droplets vary in size, waves interfere and landscapes block part of the cone. This article separates the reliable geometric mechanism from useful approximations and deeper wave effects.
- Begin with observer geometry
- Use Snell's law
- Derive the primary-bow angle
- Understand colour dispersion
- Meet the secondary bow
- Explore wave effects and limits
- Build student skills
- Read the FAQs
The Rainbow Is an Angle, Not an Object
Stand with the Sun behind you and rain in front. Draw a line from your eye directly away from the Sun; this is the antisolar direction. Light forming the primary rainbow reaches the eye from directions about 42 degrees from that line for red, with violet at a somewhat smaller angle. Rotating that direction around the antisolar axis creates a cone. The visible bow is where this cone intersects illuminated droplets.
This explains the circular form. Every suitable droplet seen at the same angular separation can contribute. Near the ground, the horizon blocks the lower part, so we usually see an arc. From a high viewpoint with rain below, a more complete circle may be visible. The circle is centred on the antisolar point, not on a particular cloud or landmark.
The geometry also explains why two people do not receive the rainbow from exactly the same droplets. Their eyes occupy different positions, so the ray paths that meet the required angle come from different droplets. They can agree that a bow appears near the same landscape direction while receiving different photons. A rainbow is observer-dependent but not imaginary; it is a repeatable optical geometry.
When the Sun is higher, the antisolar point is farther below the horizon, leaving less of the bow above the horizon. Near sunrise or sunset, the antisolar point is near the horizon and the arc can rise high. A primary bow’s top is roughly 42 degrees above the antisolar point, so if the Sun is more than about 42 degrees above a flat unobstructed horizon, the ordinary primary bow lies below the horizon for an observer at ground level.
Did You Know? The bright centre of the rainbow is not the Sun’s direction. It is opposite the Sun. If a student points one arm toward the Sun and the other toward the centre of the bow, the arms indicate nearly opposite directions. This is a safe geometry exercise only when the student does not stare at the Sun.
Refraction at a Droplet
Snell's law
When light crosses from air into water, its speed changes and its direction usually bends. Snell’s law states n1 sin i = n2 sin r. The angle i is measured from the normal in the first medium, r from the normal in the second, and n1 and n2 are refractive indices at the light’s wavelength.
The normal is a line perpendicular to the surface at the point of incidence. On a spherical droplet, the normal is the radius through that point. Angles must be measured from this radius, not from the tangent surface. Drawing the normal correctly is the most important step in a ray diagram.
For a simplified red-light example, take air index 1.000 and water index 1.331. If incidence angle i is 50 degrees, sin r = sin 50 / 1.331. Since sin 50 is about 0.7660, the ratio is about 0.5755. The inverse sine gives r about 35.1 degrees. The ray bends toward the normal because it enters the optically denser medium.
When the ray later leaves water for air, the indices reverse and the ray bends away from the normal, provided the angle allows transmission. The droplet’s spherical geometry relates the entry and exit normals, so the total deviation can be expressed using i and r. Reflection inside the droplet changes direction again.
Refractive index depends on wavelength
Water’s refractive index is not exactly the same for every visible wavelength. In normal dispersion, shorter visible wavelengths such as violet experience a slightly higher index than longer red wavelengths. Snell’s law therefore gives a slightly different refracted angle for each colour at the same incidence angle.
The difference per surface is small, but the complete path includes entry and exit refraction. Around the stationary-deviation region, those differences separate the outgoing directions enough for the eye to resolve colour bands. A rainbow is not white light being painted by the droplet; it is the spectrum being sent preferentially in different directions.
Reflection inside the water
The primary rainbow path has one internal reflection. After entering, a ray reaches the back surface of the droplet. Some energy reflects and some transmits. The reflected portion travels toward the front and refracts out. Ray diagrams often draw only the reflected route, but energy splitting is one reason not all incident light reaches the bow.
The word “reflection” does not imply that every ray undergoes total internal reflection. Fresnel reflection occurs at the boundary for a range of angles, while some light is transmitted. The fraction depends on angle, polarisation and refractive indices. A simple geometric derivation tracks direction and usually leaves intensity for later.
Geometry of the Primary Rainbow
Total deviation
Consider a ray entering a spherical droplet with incidence angle i and refraction angle r. Refraction at entry changes its direction by i – r. One internal reflection changes direction by 180 degrees – 2r in the chosen geometric accounting. Exit refraction contributes another i – r. Total deviation D from the original forward direction is therefore D = 180 degrees + 2i – 4r.
Observers usually describe the rainbow radius from the antisolar direction, which is opposite the incoming sunlight. The primary-bow angle theta is 180 degrees – D, giving theta = 4r – 2i. This formula depends on the one-reflection ray geometry and the convention for angles.
For red light near the bright rainbow ray, take i about 59.4 degrees and r about 40.2 degrees. Then theta = 4 times 40.2 – 2 times 59.4 = 160.8 – 118.8 = 42.0 degrees. The corresponding violet angle is smaller because its refractive index and stationary path differ.
Why one angle is bright
Many incidence angles strike a droplet. Their outgoing deviation changes with i, but near a particular incidence angle, the deviation reaches a stationary value: its rate of change with i is zero. A range of incoming rays is compressed into a narrow range of outgoing angles. That concentration creates a bright caustic near the rainbow angle.
Calculus makes this precise. Snell’s law links r and i. Differentiate n sin r = sin i for air index approximated as one: n cos r times dr/di = cos i. Thus dr/di = cos i divided by n cos r. For theta = 4r – 2i, dtheta/di = 4 dr/di – 2. The stationary condition sets this to zero, so dr/di = 1/2.
Combining the conditions yields 2 cos i = n cos r along with sin i = n sin r. Solving these equations gives the incidence angle associated with the geometric rainbow for that refractive index. The exact numeric result shifts with wavelength. The important point is conceptual: brightness comes from ray crowding near an extremum, not because only one special ray exists.
A numerical table
Using a fixed teaching index, students can calculate r from Snell’s law and theta = 4r – 2i for several i values. The angle rises toward a maximum and then falls. Rounding can make the top look flat, which is physically meaningful: a flat extremum means neighbouring input angles leave in similar directions.
| Incidence angle i | Refracted angle r, approximate | Rainbow radius theta, approximate |
|---|---|---|
| 50° | 35.1° | 40.4° |
| 55° | 38.0° | 42.0° |
| 59° | 40.1° | 42.4° |
| 65° | 42.9° | 41.6° |
Values depend on the refractive index and rounding, so the table is illustrative. It shows the stationary region rather than defining one exact universal radius. Atmospheric and droplet effects broaden the observed band.
Cone and circle geometry
The ray calculation gives an angular radius. To relate that angle to an apparent circle on a photograph, the camera projection must be considered. A wide-angle lens does not map equal visual angles to equal pixel distances in every projection. Measuring a rainbow radius directly on an image without lens calibration can therefore be misleading.
In three dimensions, directions at constant angle theta from an axis form a cone. Intersect a cone with a distant viewing sphere centred on the observer and the result is a circle. This is why the angular description is more fundamental than an apparent number of metres. There is no unique distance to “the rainbow”.
Why the Colours Separate
Dispersion changes the stationary angle
Because refractive index is slightly higher for violet than red in water, violet bends more strongly. When the primary path is traced, red emerges at the larger angular radius, roughly 42 degrees, and violet at a smaller radius, near 40 degrees. Looking from the centre outward, violet is on the inner edge and red on the outer edge.
The visible bands are not seven physically separated substances. The spectrum is continuous, and human colour perception groups it into named regions. Droplet size, brightness, background and eyesight affect which colours stand out. Listing seven colours can be culturally familiar while the optical distribution remains continuous.
Why the inside is brighter
For the primary bow, rays emerge over a range of angles mainly on the inner side of the caustic, while few corresponding one-reflection rays reach angles beyond the red outer edge. This contributes to a brighter region inside the primary bow and a sharper outer boundary. Scattering and background conditions also influence appearance.
The bow itself is bright because of ray concentration. Adjacent colours overlap because sunlight has finite angular size, droplets are not identical and wave effects broaden the pattern. A diagram with razor-thin monochromatic rays is an idealisation.
Sun angle and colour
At low solar elevation, sunlight passes through more atmosphere before reaching the rain and observer. Shorter wavelengths can be more strongly scattered out of the direct solar beam, making a rainbow appear redder near sunset. This is atmospheric filtering before and after the droplet, not a reversal of water’s dispersion.
Photographs can exaggerate or suppress colours through exposure, white balance, polarising filters and image processing. A polariser can change rainbow contrast because reflection and refraction produce polarisation patterns. A photograph is data from an optical system, not an untouched copy of the sky.
For a broader mathematical treatment of digital colour, read Why Mathematics? | Colour Spaces, Gamma Correction and Digital Displays. For controlled light mixing, see Why Mathematics? | Stage Lighting, Inverse-Square Law and Additive Colour Mixing.
Secondary Rainbows and the Dark Band
Two internal reflections
A secondary rainbow forms from rays that reflect twice inside droplets before leaving. Each reflection changes the geometry and loses additional energy, so the secondary bow is generally fainter. Its angular radius is larger, roughly around 50 to 53 degrees from the antisolar direction, depending on wavelength and conditions.
For two internal reflections, a useful deviation expression is D2 = 360 degrees + 2i – 6r under one convention. Converting to the observed direction requires care because the outgoing ray lies on a different branch. Students should derive from a labelled diagram rather than blindly change coefficients. The reliable visible result is a larger-radius, fainter bow with reversed colour order.
In the secondary bow, red lies on the inner edge and violet on the outer edge. The reversal follows from how the two-reflection deviation varies with refractive index. A physical mnemonic is helpful only after the ray diagram; otherwise students may remember the order without understanding the path.
Alexander's dark band
The region between the primary and secondary bows can appear darker than the sky inside the primary or outside the secondary. One-reflection rays contribute mainly inside the primary, while two-reflection rays contribute mainly outside the secondary. The angular gap receives fewer of those rainbow rays. This is Alexander’s dark band.
The band is not a dark material in the atmosphere. It is a direction with less added droplet light under suitable conditions. Background cloud brightness, rainfall and sunlight influence how obvious it looks. Its appearance is another clue that the bow is an angular distribution rather than a painted object.
Higher-order bows
Rays can reflect three or more times inside droplets, producing higher-order rainbows. They are much fainter and may occur toward or away from the Sun depending on order. Additional reflection loss and bright sky background make them difficult to see. Special observation and image processing have revealed some higher orders, but they are not expected in every shower.
Mathematically, each extra reflection adds a geometric term and changes the stationary-deviation condition. The pattern illustrates how one physical mechanism can generate a family of solutions. It also shows why model predictions may exist long before an everyday observer can verify them easily.
NOAA’s What Causes a Rainbow? gives an accessible official explanation of refraction, reflection and dispersion. Its Rainbow Simulator can support safe exploration of Sun-observer geometry.
Where the Ray Model Needs Help
Droplets are not perfect in every situation
Small freely falling water droplets are close to spherical because surface tension dominates. Larger raindrops flatten and distort as aerodynamic forces increase. A distorted drop changes ray geometry and can alter the lower parts of a bow. The spherical model remains an excellent first approximation, but not a law about every drop shape.
Droplet size affects sharpness. A fine mist can produce a broad, pale fogbow because diffraction and overlapping colours become important. Larger, more uniform droplets can yield vivid colours and supernumerary bands. A purely geometric ray has no wavelength-scale width, so wave optics is needed to explain these details.
Supernumerary bows
Sometimes faint pastel bands appear just inside the primary rainbow. These supernumerary bows arise from interference between light paths with nearly the same direction but different optical path lengths. Where waves arrive in phase, intensity is enhanced; where out of phase, it is reduced.
Geometric optics predicts the main caustic but not the full interference pattern. Wave analysis near a caustic uses more advanced mathematics, including Airy-type functions. Students do not need that machinery to appreciate the lesson: models have domains, and a new observation can reveal which omitted feature matters.
The Sun is not a point
The Sun spans about half a degree in the sky. Rays arrive from slightly different directions across its disc, blurring sharp angular features. Droplet-size distributions, multiple scattering and detector response add further smoothing. Therefore quoting a rainbow angle to many decimal places would misrepresent natural variability and measurement conditions.
Polarisation and intensity
Fresnel equations describe how reflection and transmission amplitudes depend on angle and polarisation. Rainbow light is strongly polarised in characteristic directions. This is why a rotating polarising filter can change contrast. Directional geometry locates the bow, while electromagnetic boundary conditions help predict its intensity.
Atmosphere and landscape
The ray model assumes direct sunlight reaches droplets and the scattered light reaches the observer. Clouds can block the Sun, terrain can hide part of the cone and haze can reduce contrast. A nearby spray may create a small-looking bow because the droplets occupy only part of the required cone, but the angular radius remains governed by optics.
NASA’s Astronomy Picture of the Day entry A Horizon Rainbow in Paris illustrates how observer position and the horizon shape the visible arc. The image is a useful prompt for separating angular geometry from apparent landscape placement.
Measuring a rainbow photograph
A photograph can support a careful project if its limitations are documented. First identify or estimate the antisolar point. If the horizon is level and the Sun’s elevation is known, the antisolar point lies the same angle below the opposite horizon. Next trace several points along one colour edge and fit a circle in image coordinates. The fitted centre should be consistent with the projected antisolar direction.
Pixels are not degrees until the lens projection is known. A rectilinear lens maps angle through a tangent function, while fisheye projections use different relationships. Cropping removes information about the optical centre. Students should therefore report a pixel-radius fit as an image measurement and convert to angle only with calibration or reliable metadata.
Uncertainty can be estimated by repeating the fit with points chosen by different observers or along different colour boundaries. If the centre and radius change substantially, the edge is ambiguous or the projection model is poor. Residuals—the distances between traced points and fitted circle—can reveal lens distortion, perspective assumptions or a bow partly hidden by cloud.
This project connects geometry, least-squares fitting and scientific communication. A beautiful image becomes data, but the conclusion remains proportional to the calibration. The strongest report may say “consistent with a circular arc” rather than claim an exact 42.000-degree radius from an uncalibrated phone photograph.
From Fermat's principle to the bow
An advanced student can connect the stationary rainbow angle with a broader optical idea: nearby paths around a stationary optical condition have very similar phase or travel-time behaviour. Fermat’s principle is often introduced as light taking a stationary optical path. Snell’s law can be derived from that principle at a flat boundary.
The rainbow caustic is not simply the shortest path between two fixed points, so the concepts should not be collapsed carelessly. The useful connection is stationary behaviour. When an output changes very little as an input varies, many neighbouring contributions can gather. Similar mathematics appears in focusing, envelopes, optimisation and asymptotic wave analysis.
Students can recognise the pattern without advanced calculus. Plot a smooth hill and note that horizontal position changes while height changes little near the top. The primary rainbow’s angular curve has a comparable flat region near its extremum. Calculus names this with a zero derivative; observation reveals it as concentrated brightness.
How Students Can Learn This Mathematics
Draw normals first
On a large circle, mark an entry point and draw the radius to it. That radius is the normal. Draw the incoming ray and label incidence angle from the radius. Refract toward the normal, continue to the back surface, reflect with equal incident and reflected angles, and refract out. Only then add i, r and deviation labels.
Many mistakes vanish when normals are correct. Measuring from the tangent instead of the normal replaces an angle by its complement and breaks Snell’s law. Colour-code the ray, radius and tangent so the roles remain distinct.
Calculate a Snell table
Choose a clearly labelled teaching index for red and another for violet. For incidence angles from 45 to 70 degrees, calculate r = arcsin(sin i/n) and then theta = 4r – 2i. Plot theta against i for both indices. Find the maximum of each curve and compare angular radii.
Students should set calculators to degrees and verify that inverse sine receives a number between -1 and 1. A radians-versus-degrees error creates plausible-looking nonsense. Test i = 0: r should be 0. Although that ray does not form the bright bow, it is a clean formula check.
Use geometry safely outdoors
Observe a natural rainbow only with the Sun behind you. Never stare at the Sun or point unfiltered optical instruments toward it. Identify the antisolar direction from shadows rather than looking back at the solar disc. Estimate the bow’s angular size using a safe familiar angular reference, recognising that hand-span rules vary by person and arm length.
A garden-hose mist can show a bow when local water rules, footing and adult supervision make it safe. Keep water away from roads, electrical equipment and slippery surfaces. The goal is to change observer and spray position and notice the angular condition, not to maximise spray.
Build a simple simulation
A spreadsheet can implement Snell’s law and the primary deviation formula. Create columns for i, sin i, r and theta. Use small incidence increments and a chart. Then vary refractive index slightly. The maximum shifts, and the gap between colour curves becomes visible.
A more advanced coding project can trace rays through a circle using vectors. Find intersections, construct surface normals, apply vector reflection and Snell refraction, and record exit direction. Comparing numerical output with the analytic formula tests both approaches. The program should conserve unit-vector length and flag total internal reflection conditions appropriately.
Add automated tests before trusting the picture. A ray aimed through the centre should meet both surfaces normally. Reflection should preserve the angle to the normal. Reversing a refracted path should retrace it when the media are swapped. These checks use symmetry and reversibility as mathematical evidence. A colourful plot is useful only after the underlying geometry passes such tests.
Transferable problem-solving routine
- Define the observer, light source and object geometry.
- Draw coordinate axes and surface normals.
- State refractive indices and wavelength assumptions.
- Apply Snell’s law at each crossing.
- Account for each reflection and direction change.
- Plot output angle against input angle.
- Locate stationary behaviour and compare colours.
- State where wave optics and real conditions matter.
This routine transfers to lenses, prisms, fibre optics and imaging. The recurring habits are careful geometry, consistent angle conventions and model checking.
A Guide for Parents and Teachers
Begin with the observer. Ask: where is the Sun, where is the rain and which direction is opposite the Sun? This anchors the geometry before equations. A student who understands the antisolar axis can explain the arc, centre and changing height.
Use clear language about models. “Spherical droplet and one internal reflection” should appear beside the primary-bow calculation. Later, show a fogbow or supernumerary image and ask which assumption or omitted wave behaviour becomes important. This turns exceptions into learning rather than confusion.
Separate three achievements. First, the student can use Snell’s law numerically. Second, the student can assemble the droplet geometry. Third, the student can explain why a stationary angle is bright. Mastery grows in layers; calculus should deepen the geometric story rather than obscure it.
Encourage estimation. If a calculation gives a primary radius of 4 degrees or 140 degrees, it fails a visual reality check. If refraction into water bends away from the normal, the direction is wrong. Physical sense should accompany calculator accuracy.
Keep observation safe. Never ask students to face or photograph the Sun directly with magnifying optics. Use shadows to establish solar direction, public simulations for controlled variation and professionally prepared images for detailed measurement.
Career connections include physics, atmospheric science, optical engineering, remote sensing, imaging, colour science and computer graphics. Mathematics is valuable preparation but guarantees no course or job. Students should build experiments, coding and communication alongside algebra and trigonometry, and verify current pathway requirements directly.
Common Misconceptions
“A rainbow is located at one distant place.” It is an angular optical phenomenon relative to each observer, Sun and droplets.
“Only one ray enters each droplet.” Many rays enter. Brightness concentrates near a stationary-deviation direction.
“The primary bow uses two reflections.” It uses one internal reflection; the secondary uses two.
“The seven colours are seven separate beams.” The spectrum is continuous, though dispersion and perception create recognisable bands.
“Red bends more than violet in water.” Under normal visible dispersion, violet has a higher refractive index and bends more at entry; primary-bow geometry places violet on the inner edge.
“A full circular rainbow needs circular rain.” The direction cone is circular around the antisolar axis. A high viewpoint and illuminated droplets below can reveal more of it.
“Ray optics explains every detail.” It explains the main bow and caustic. Interference, diffraction, polarisation and drop deformation require richer models.
Frequently Asked Questions
Why must the Sun be behind the observer?
Primary rainbow light returns after refraction, one internal reflection and exit refraction. The outgoing directions form a cone around the line opposite the Sun.
Why is the primary rainbow about 42 degrees?
For water’s visible refractive index, the one-reflection ray deviation has a stationary value. Red light appears near a roughly 42-degree radius from the antisolar direction; other wavelengths differ slightly.
Why is red on the outside?
Water’s refractive index varies with wavelength. The stationary path for red produces a larger primary-bow angular radius than violet, so red is farther from the antisolar centre.
Why are secondary rainbows fainter?
They involve two internal reflections, so more light is lost through transmission and reflection at boundaries. Their light is also spread over a different angular region.
Why is the secondary colour order reversed?
The extra reflection changes how deviation varies with refractive index. The larger secondary cone places red on the inner edge and violet on the outer edge.
Can there be a rainbow without rain?
Yes. Any suitable water droplets, such as mist, spray or fog, can create related bows when illumination and observer geometry are correct.
Why do some rainbows have extra inner bands?
Supernumerary bands result from wave interference between nearby optical paths. Their visibility depends strongly on droplet size and uniformity.
What mathematics should students know first?
Angles, circles, trigonometric functions, inverse sine, graphs and algebra are enough for the basic model. Calculus explains the stationary angle, while wave mathematics explains finer bands.
Is the rainbow exactly 42 degrees?
No single exact angle covers every wavelength and condition. Refractive index varies with colour and temperature, the Sun has angular size, and droplets and wave effects broaden the bow. About 42 degrees is a useful primary-red approximation.
Useful Next Reading
- NOAA NESDIS: What Causes a Rainbow? gives an official, accessible optical explanation.
- NOAA NESDIS: Rainbow Simulator supports safe exploration of the Sun-observer-droplet geometry.
- NASA APOD: A Horizon Rainbow in Paris provides an observed example for analysing angular placement.
- Why Mathematics? | Colour Spaces, Gamma Correction and Digital Displays explains how devices encode and display colour.
- Why Mathematics? | Stage Lighting, Inverse-Square Law and Additive Colour Mixing connects light intensity, geometry and colour mixing.
A Final Thought
Rainbows show why mathematics is important in a form anyone can see. Trigonometry follows a ray through a droplet. Geometry turns a fixed angle into a circular cone. Calculus finds the caustic where rays crowd. Dispersion separates wavelengths. Wave theory explains delicate bands that the first model cannot. Each layer answers a question and reveals a new one.
The best mathematical understanding keeps the sky and the equation together. It checks angle conventions, respects approximations and remains alert to observation. A rainbow is neither a simple painted arc nor an incomprehensible miracle. It is a rich meeting of sunlight, water, position and mathematical structure—and learning that structure gives students another reason to look carefully at the world.