08 October 2026

What Actually Happens When a Cold Front Arrives?

 


What Actually Happens When a Cold Front Arrives?

Look at almost any television weather forecast and sooner or later you will see a line moving across the map decorated with triangles or semicircles.

The presenter may say:

“A cold front will move across the country during the afternoon, bringing a band of rain.”

But what actually is a cold front?

It is tempting to imagine it as some sort of invisible wall travelling across the landscape. Textbook diagrams do not always help. They often show a neat wedge of cold air sliding underneath warm air, with clouds conveniently appearing above it.

The real atmosphere is considerably more complicated.

However, there is a wonderfully simple experiment that can make one of the most important ideas visible.

Instead of trying to watch two invisible masses of air collide, we can use warm and cold coloured water in a transparent tank.

Suddenly, density becomes something we can actually see.


What Is a Weather Front?

A front is essentially a boundary between two air masses with different properties.

Those differences might include:

  • temperature;

  • humidity;

  • density;

  • origin;

  • and sometimes wind direction.

An air mass that has spent time over a cold region can be considerably colder than one arriving from a warmer region.

When those air masses meet, they do not necessarily mix instantly.

Their different densities matter.

And that gives us our experiment.


The Tank Experiment

For a simple demonstration, I can use a transparent tank containing bodies of water at different temperatures.

The cold water can be coloured blue.

The warm water can be coloured red.

Ideally, the colours should make the boundary between them easy to see.

The important question is:

What happens when the two fluids meet?

Before doing anything, I would ask students to predict the result.

Will they:

  1. mix immediately?

  2. remain completely separate?

  3. have the warm water move underneath the cold?

  4. have the cold water move underneath the warm?

Making the prediction first turns a demonstration into an investigation.


Watch the Blue Water

As the cold and warm water meet, something interesting should become visible.

The colder water tends to move beneath the warmer water.

Why?

Because temperature affects density.

For most everyday conditions, cooling water makes it denser.

A given volume of colder water therefore tends to contain slightly more mass than the same volume of warmer water.

Under gravity, the denser fluid tends to sink beneath the less dense fluid.

That is the first important connection with weather.

Cold air is generally denser than warm air.

So when a mass of colder air advances into a region occupied by warmer air, the cold air can push underneath it.

The warm air is forced upwards.

That upward movement is crucial.

Because the interesting weather often happens not simply because the cold air has arrived, but because of what happens to the warm air that is lifted above it.


Why Rising Air Matters

Suppose warm, moist air is sitting close to the ground.

A cold front approaches.

The denser cold air begins moving underneath the warmer air.

The warm air is pushed upwards.

As that air rises, atmospheric pressure decreases.

The rising air therefore expands.

When a gas expands under these conditions, its temperature falls.

This is known as adiabatic cooling.

Eventually, the rising air may cool to its dew point.

Water vapour then begins condensing onto tiny particles in the atmosphere called condensation nuclei.

Tiny water droplets form.

A cloud begins to develop.

Continue lifting sufficiently moist air and those droplets may grow large enough to produce precipitation.

So the sequence is approximately:

cold air advances → warm air rises → air expands → air cools → condensation occurs → clouds develop → precipitation may follow

That is a much more satisfying explanation than simply memorising:

“Cold fronts bring rain.”


Why Cold Fronts Can Produce Dramatic Weather

Cold fronts often have a relatively steep boundary.

The advancing cold air can force warm air upwards quite rapidly.

Rapid uplift can encourage strong convection and substantial vertical cloud development when the atmosphere contains sufficient moisture and is unstable.

This is why an active cold front can sometimes be associated with:

  • towering cumulonimbus clouds;

  • heavy showers;

  • sudden downpours;

  • squally winds;

  • thunderstorms;

  • hail.

But this is also where we should be careful with school-level simplifications.

A cold front does not automatically produce a thunderstorm.

The resulting weather depends upon many factors, including the amount of moisture present, atmospheric stability, temperature structure and the dynamics of the weather system.

Sometimes a cold front produces dramatic weather.

Sometimes its passage is much less impressive.

That uncertainty is part of real meteorology.


What About a Warm Front?

Now consider the opposite situation.

Instead of dense cold air advancing beneath warm air, imagine warmer air advancing towards an existing mass of colder air.

The warm air is less dense and cannot simply bulldoze the cold air out of the way.

Instead, it tends to rise over it.

The slope associated with a warm front is typically much gentler than that of a cold front.

That means the uplift can occur gradually over a much greater horizontal distance.

This helps explain why warm fronts are often associated with extensive layers of cloud and prolonged precipitation rather than the narrower, sometimes more intense weather associated with cold fronts.

It also explains why the weather can begin changing well before the warm front itself reaches an observer.

High cloud may appear first.

It may gradually thicken and lower.

Eventually rain can arrive.

The changing sky is effectively revealing what is happening to the air many kilometres away.


Can We Investigate This Rather Than Just Demonstrate It?

This experiment becomes much more interesting if we start changing variables.

Experiment 1 — Change the Temperature Difference

Try two situations.

Small difference:
Cold and warm water only a few degrees apart.

Large difference:
A considerably greater temperature difference.

Does the movement of the fluids change?

Does the colder fluid penetrate underneath the warmer fluid more obviously?

How quickly does mixing occur?

This introduces the idea that the magnitude of a density difference matters.


Experiment 2 — Change the Rate of Introduction

Introduce the colder water very slowly.

Repeat while introducing it more rapidly.

The behaviour may be noticeably different.

A gentle introduction can produce a clearer boundary.

A faster introduction may generate considerably more turbulence and mixing.

That immediately gives us another useful lesson.

Real atmospheric fronts are not perfectly smooth surfaces.

They contain turbulence, eddies and complex three-dimensional movements.


Experiment 3 — Change the Shape of the Boundary

Instead of allowing the fluids to meet across a vertical boundary, try arranging the experiment so that the interface begins at an angle.

Watch how the boundary evolves.

Film it from the side.

Better still, record it and play it back slowly.

Small-scale movements that are difficult to notice during the demonstration can become much more obvious on video.


Turn the Experiment Into Measurements

We can go further than simply saying:

“Look — the cold water goes underneath.”

Place a scale behind the tank.

Record the experiment from a fixed camera position.

Measure how far the leading edge of the cold water travels at regular time intervals.

For example:

Time: 0 s, 5 s, 10 s, 15 s, 20 s...

Position: measured from the video.

Now plot:

distance against time

Repeat the experiment using different temperature differences.

We have moved from a colourful demonstration into a quantitative investigation.

Students can begin asking genuine scientific questions:

Does increasing the temperature difference increase the speed at which the denser fluid moves beneath the warmer fluid?

That is much closer to real experimental science.


A Thermal Camera Could Add Another Dimension

An especially interesting extension would be to compare the coloured-water view with thermal imaging.

The dye shows where the fluids move.

A thermal camera may reveal how the temperature distribution changes.

Those are not necessarily identical things.

As mixing occurs, colour boundaries and temperature boundaries may evolve differently.

That creates another useful discussion:

What exactly is each instrument measuring?

Scientific instruments do not simply give us “the answer”.

Each gives us a particular type of information.

Combining observations often gives us a much better understanding of the system.


But Water Is Not Air

This is perhaps the most important part of the experiment.

The tank is not a miniature atmosphere.

Water is a liquid.

Air is a compressible gas.

The atmosphere exists on an enormous scale.

Earth rotates.

Pressure varies with altitude.

Humidity matters.

Solar heating matters.

Terrain matters.

Large-scale pressure systems matter.

The atmosphere also contains turbulent motion on scales ranging from centimetres to hundreds or thousands of kilometres.

Our tank contains almost none of that.

So why use it?

Because models do not need to reproduce every feature of reality to be useful.

They need to isolate something important.

In this case, the model helps us visualise the behaviour of fluids with different densities under gravity.

That is a genuine physical principle involved in atmospheric behaviour.

Understanding where a model works and where it fails is arguably more scientifically valuable than pretending the model is perfect.


Another Important Difference — Water Has Its Own Peculiarities

There is another reason to be careful.

Water does not simply become denser indefinitely as it cools.

Fresh water reaches its maximum density at approximately 4°C. Below this, its behaviour becomes unusual, which ultimately helps explain why ice floats.

So if this experiment is designed specifically to illustrate the ordinary relationship between warmer and colder fluids, there is no need to use near-freezing water.

Moderately cold and warm water will make the point perfectly well and avoids introducing an unnecessary complication.

Of course, that unusual behaviour of water could become an excellent separate investigation.

One simple experiment can generate several new scientific questions.


From the Tank to the Weather Map

Once students have watched the experiment, the familiar weather symbols become much more meaningful.

A cold front is usually shown as a line with triangles pointing in the direction of movement.

A warm front uses semicircles.

An occluded front combines the symbols.

But now those lines are no longer merely marks to memorise.

They represent boundaries between air masses.

Behind those symbols lies fluid dynamics.

There are differences in temperature.

Differences in density.

Vertical motion.

Cooling.

Condensation.

Cloud formation.

Rainfall.

And sometimes dramatic changes in wind and weather.


Could You Detect a Cold Front Passing Your Own Home?

This creates a fascinating follow-up investigation.

If you have access to a weather station, watch what happens as a forecast cold front passes.

Record:

  • air temperature;

  • atmospheric pressure;

  • wind speed;

  • wind direction;

  • rainfall;

  • relative humidity.

Then compare those measurements with the published weather charts.

Can you identify approximately when the front passed?

Perhaps the temperature falls.

Perhaps the wind changes direction.

Perhaps there is a period of rain.

Perhaps pressure begins to rise behind the front.

Suddenly, the neat coloured lines on a national weather map connect directly with measurements made in your own garden.

That is where meteorology becomes particularly engaging.

We stop merely watching the weather forecast.

We start testing it against observations.


A Front Is Not Really a Line

There is one final misconception worth challenging.

On a weather map, a front is drawn as a line.

In reality, it is not an infinitely thin boundary.

It is a three-dimensional transition zone extending vertically through the atmosphere and horizontally across a substantial distance.

The line on the map is therefore another model.

It takes an enormously complicated three-dimensional atmospheric structure and represents it with a simple two-dimensional symbol.

Once again, simplification is useful.

But we should always remember that the atmosphere itself is much more complicated than the diagram.


Why I Like This Experiment

I particularly like demonstrations in which something normally invisible suddenly becomes visible.

We cannot easily watch one air mass sliding beneath another.

We cannot see density.

We cannot directly see atmospheric uplift.

But with a transparent tank, two temperatures of water and a little colouring, we can make an analogous process visible enough to investigate.

Then we can move from the tank to the atmosphere.

From density to fronts.

From fronts to uplift.

From uplift to cooling.

From cooling to condensation.

And from condensation to the clouds and rain we see outside.

That progression is what makes a simple experiment scientifically powerful.

Weather Fronts in a Tank

Modelling What Happens When Warm and Cold Air Masses Meet

Aim

To investigate what happens when two fluids of different temperatures — and therefore different densities — meet.

The experiment models one important feature of atmospheric fronts:

colder, denser fluid tends to move underneath warmer, less dense fluid.

It can then be used to introduce:

  • cold fronts;

  • warm fronts;

  • density currents;

  • uplift;

  • convection;

  • cloud formation;

  • rainfall;

  • the strengths and limitations of scientific models.


Equipment

You will need:

  • long transparent tank or aquarium;

  • removable vertical divider that fits reasonably closely across the width of the tank;

  • two large measuring jugs;

  • thermometers or temperature probes;

  • warm water;

  • cold water;

  • blue food colouring;

  • red or yellow food colouring;

  • stopwatch;

  • ruler or measuring scale;

  • white background card;

  • camera or phone on a tripod;

  • optional temperature probes or data logger;

  • optional thermal camera;

  • towels or absorbent cloths.

A tank approximately 60–100 cm long is ideal, although a smaller tank will still work.

A relatively shallow rectangular tank is often better than a very deep aquarium because the horizontal movement is easier to see.


Recommended Temperatures

You do not need boiling water or ice water.

A good starting point is:

Cold side: about 10–15°C

Warm side: about 30–35°C

That gives a large enough temperature difference for the density effect to be visible while remaining straightforward to handle.

For a more quantitative investigation, repeat using smaller differences such as:

  • 15°C and 25°C;

  • 15°C and 30°C;

  • 15°C and 35°C.

Avoid making the cold water extremely close to 0°C, because the unusual density behaviour of water near 4°C can complicate the interpretation.


Preparing the Tank

Place the empty tank on a level bench.

Fix a white sheet or white card behind it. This greatly improves visibility of the coloured water.

Attach a horizontal measuring scale to the outside of the tank.

If possible, mark distances every 5 cm.

Place the removable divider vertically across the centre of the tank.

The divider needs to separate the tank into two compartments.

It does not need to be completely watertight for a classroom demonstration, but the closer the fit, the cleaner the start of the experiment.

A sheet of:

  • acrylic;

  • thin polycarbonate;

  • plastic sheet;

  • or rigid laminated card

can work well.


Preparing the Two Water Masses

Prepare equal volumes of water.

For example:

Left-hand side:
3 litres warm water at approximately 35°C.

Add a few drops of red food colouring.

Right-hand side:
3 litres cold water at approximately 12°C.

Add blue food colouring.

Use only enough dye to identify the fluids clearly.

Too much food colouring makes the mixture so dark that the boundary becomes difficult to see.

Record the actual temperatures.


Filling the Tank

This part should be done fairly carefully.

First pour the warm coloured water into one side of the divider.

Then pour the cold coloured water into the other.

Try to keep the water levels approximately equal.

If the levels are different, hydrostatic pressure will produce a flow when the divider is removed, and this could be mistaken for a density effect.

Allow the water to settle for approximately 30–60 seconds.

Record the temperature on both sides immediately before the experiment starts.


Prediction

Before removing the divider, ask:

What do you think will happen when the divider is lifted?

Possible predictions include:

  • both fluids immediately mix;

  • the cold water travels underneath the warm water;

  • the warm water travels underneath the cold water;

  • the two remain separated;

  • one fluid rises while the other sinks.

Students should record their prediction before seeing the result.


Performing the Experiment

Start recording the experiment with the camera.

The camera should ideally be:

  • side-on to the tank;

  • level with the centre of the tank;

  • fixed on a tripod;

  • far enough away to show the whole tank.

Start the stopwatch.

Lift the divider vertically upwards in one smooth movement.

Do not pull it sideways.

Do not remove it excessively quickly, because that can generate unnecessary turbulence.

Observe the coloured water.


What You Should See

The cold blue water should begin travelling underneath the warmer red water.

Near the bottom of the tank, the blue water forms a spreading density current.

At the same time, some of the warmer fluid is displaced upwards.

The boundary between the two fluids may slope.

You will also see:

  • rolling motion;

  • eddies;

  • mixing;

  • turbulence;

  • coloured structures developing along the interface.

The blue and red regions will eventually mix, but the early part of the experiment should make the density difference quite obvious.


What Is Happening?

The colder water is slightly denser than the warmer water.

When the divider is removed, gravity allows the denser fluid to move beneath the less dense fluid.

The warm fluid is displaced upwards.

This behaviour gives us a useful analogy for a cold atmospheric air mass advancing beneath warmer air.

The important idea is:

cold dense fluid underneath — warm less dense fluid above

In the atmosphere, that vertical displacement can force warm moist air upwards.


Connecting the Tank to a Cold Front

A cold front occurs when colder air advances into a region containing warmer air.

The colder air is usually denser.

It therefore tends to push beneath the warmer air.

That forces the warmer air upwards.

As the warm air rises:

  1. atmospheric pressure decreases;

  2. the air expands;

  3. its temperature falls;

  4. relative humidity increases;

  5. eventually saturation may occur;

  6. water vapour condenses;

  7. clouds develop;

  8. precipitation may follow.

The tank demonstrates the first part of this sequence particularly well:

dense cold fluid pushing underneath less dense warm fluid.


Measuring the Density Current

The experiment can easily become quantitative.

Choose the blue cold-water front as the feature to track.

From the video, record the position of the leading edge at regular intervals.

For example:

Time / sDistance travelled / cm
00
2
4
6
8
10
12

Plot:

distance travelled against time

You could also calculate an approximate velocity:

velocity = distance travelled / time

Repeat the experiment with different temperature differences.

Then ask:

Does a greater temperature difference produce a faster density current?


Investigation 1 — Temperature Difference

Keep everything else the same.

Try:

Trial A

Cold: 15°C
Warm: 20°C

Trial B

Cold: 15°C
Warm: 30°C

Trial C

Cold: 15°C
Warm: 40°C

Measure how quickly the cold-water front travels.

The independent variable is:

temperature difference

The dependent variable could be:

speed of the cold-water front

Control variables should include:

  • volume of water;

  • tank dimensions;

  • dye concentration;

  • starting water depth;

  • divider position;

  • method of removing the divider.


Investigation 2 — Does Water Depth Matter?

Repeat using different depths of water.

For example:

  • 5 cm;

  • 10 cm;

  • 15 cm.

Measure the speed and appearance of the cold-water current.

Students can investigate whether the geometry of the fluid changes the behaviour.


Investigation 3 — Introduce One Fluid Gradually

Instead of using a divider, begin with warm water in the tank.

Carefully introduce cold blue water at one end near the bottom.

A length of tubing can help introduce it gently.

The cold water should spread along the bottom.

Repeat by introducing warm coloured water into cold water near the top.

Compare the two situations.

This version can make the difference between an advancing cold current and an overrunning warm fluid particularly clear.


Investigation 4 — Model a Cold Front and a Warm Front

You can run two versions.

Cold-front model

Start with warm water occupying most of the tank.

Introduce cold blue water from one end at low level.

Observe the denser fluid moving underneath.

This resembles the basic geometry of an advancing cold front.

Warm-front model

Begin with colder water occupying most of the tank.

Introduce warm red water gently near the surface.

The warmer, less dense fluid tends to remain above the colder water.

This provides a simple analogy for warm air overrunning a colder air mass.

The analogy is not perfect, but the contrast between the two experiments is educationally very useful.


Investigation 5 — Add Temperature Probes

If temperature probes are available, place them at different positions.

For example:

  • bottom left;

  • middle;

  • bottom right;

  • near the surface.

As the cold current moves through the tank, record the temperature at each location.

This makes it possible to watch a simulated “front” pass a fixed point.

That is particularly interesting because it can then be compared with what a weather station records when a real atmospheric front passes.

At a fixed point, the temperature may suddenly change as one fluid replaces another.


A Strong Weather-Station Connection

Once students have seen the tank experiment, show them data from a real frontal passage.

Look for changes in:

  • temperature;

  • atmospheric pressure;

  • wind direction;

  • wind speed;

  • humidity;

  • rainfall.

You can then ask:

What is the tank showing that the weather station cannot show directly?

and:

What does the weather station measure that the tank does not model?

This is a very good way of discussing the difference between a laboratory model and the real atmosphere.


Using a Thermal Camera

A thermal camera could make this particularly impressive.

Film the experiment normally from one side.

Then examine the tank thermally.

The visible-light image shows the coloured fluids.

The thermal image shows the temperature distribution.

You may find that the colour boundary and temperature boundary gradually become less sharply aligned as mixing takes place.

This leads to an excellent question:

Does the colour show temperature, or does it merely show where the original water came from?

The answer is that the dye is a tracer.

It identifies fluid movement.

It is not itself a temperature measurement.

That distinction is important experimental science.


A More Dramatic Version

For filming, I would use:

  • blue cold water;

  • amber or red warm water;

  • strong white backlighting;

  • a black or darkened laboratory around the tank;

  • a fixed close-up side camera;

  • a second camera looking slightly downwards;

  • slow-motion recording if available.

The tank can look remarkably atmospheric as the coloured fronts roll over one another.

A ruler attached to the back also ensures that the demonstration remains visibly scientific rather than becoming simply a colourful effect.


Expected Observation

The main observation should be:

The cold, denser water moves beneath the warmer, less dense water.

This is a form of a gravity current or density current.

A similar physical principle contributes to the behaviour of air masses.


Important Limitation

This experiment does not literally reproduce a weather front.

The atmosphere differs from the tank because:

  • air is a gas rather than a liquid;

  • air is compressible;

  • atmospheric pressure changes considerably with height;

  • the Earth rotates;

  • the atmosphere is continually heated and cooled;

  • humidity affects cloud formation;

  • wind exists in three dimensions;

  • terrain affects airflow;

  • atmospheric fronts can extend for hundreds or thousands of kilometres.

The tank models one central principle:

fluids of different densities tend to arrange themselves with the denser fluid beneath the less dense fluid.

That principle helps explain why advancing cold air can push underneath warmer air.


Questions for Students

  1. Which fluid moved closest to the bottom of the tank?

  2. Why did it do this?

  3. How did increasing the temperature difference change the result?

  4. Why was the interface between the fluids not perfectly smooth?

  5. What atmospheric process is represented by the warm fluid being displaced upwards?

  6. Why can rising warm moist air produce clouds?

  7. Why is this experiment only a model of a weather front?

  8. What variables would have to be controlled to compare two experiments fairly?

  9. How could the movement of the front be measured quantitatively?

  10. What would you expect a weather station to record as a real cold front passed?


Suggested Conclusion

The experiment demonstrates that two fluids at different temperatures do not necessarily mix immediately.

Because the colder fluid is denser, it tends to move underneath the warmer fluid.

This creates a visible density current.

In the atmosphere, colder air can similarly push beneath warmer air at a cold front.

The resulting uplift of warm moist air can cause cooling, condensation, cloud formation and rainfall.

The tank is therefore not a miniature atmosphere, but it provides a powerful model of one of the physical principles that helps make weather fronts behave as they do.


The Bigger Lesson

The next time a weather forecast says:

“A cold front will move through this afternoon…”

do not imagine a mysterious line travelling across the country.

Imagine an enormous three-dimensional interaction between air masses.

Imagine denser cold air advancing.

Imagine warmer air being lifted.

Imagine that rising air expanding and cooling.

Imagine microscopic droplets beginning to form.

And then look at the clouds.

The atmosphere is performing an enormous fluid-dynamics experiment above our heads every day.

The coloured water in the tank does not reproduce all of it.

But it gives us a window into the physics that makes weather happen.

Sometimes the best way to understand something as enormous as the atmosphere is to begin with something small enough to put on the laboratory bench.


07 October 2026

Knot Theory — When Is a Knot Really the Same Knot?

 


Knot Theory — When Is a Knot Really the Same Knot?

Anyone can tie a knot. Proving which knot you have tied is considerably harder.

Give a student a piece of string and ask them to tie a knot.

It sounds like something for a five-year-old rather than a mathematics lesson.

Now give them two pieces of string, tie a different-looking knot in each, and ask:

Are these actually different knots?

That question is considerably harder.

You can pull them. Twist them. Stretch them. Turn them over. Move one loop through another.

But there is one important rule:

You are not allowed to cut the string.

Suddenly, playing with string has become mathematics.

Welcome to knot theory, a branch of topology in which mathematicians try to understand knots, links and tangled structures.

And one of its most fundamental questions is wonderfully simple to state:

How can we prove that two knots are different?


First, There Is a Problem With Ordinary String

If I take a normal piece of string, tie a knot in it and leave the two ends free, I can usually undo the knot by manipulating one of the ends.

Mathematicians therefore do something slightly different.

Imagine joining the two ends of the string together.

We now have a closed loop.

If the loop contains no knot, we have the simplest possible mathematical knot:

the unknot.

It is essentially just a loop.

Now suppose we tie a trefoil knot in the string before joining the ends.

We have produced something fundamentally different.

The question is:

How do we know?

It might look different, but appearances can be deceptive.


Topology — Mathematics Without a Ruler

Knot theory belongs to a larger area of mathematics called topology.

Topology is sometimes described informally as the mathematics of shape when distances and angles do not matter.

Imagine an object made from perfectly flexible rubber.

You are allowed to:

  • stretch it;

  • squash it;

  • bend it;

  • twist it.

But you cannot:

  • tear it;

  • cut it;

  • glue previously separate pieces together;

  • pass one part magically through another.

Under those rules, many objects that look different geometrically are actually regarded as the same topological object.

This is a very different way of thinking from the geometry students normally meet at school.

In geometry, changing a square into a rectangle matters because the lengths and angles have changed.

In topology, many such changes are irrelevant.

What matters is the underlying structure.

And that is exactly the idea we need for knots.


Experiment 1 — Can You Fool Someone With the Unknot?

Start with a simple closed loop of cord.

Now twist and distort it.

Lay it on a table so that one part crosses another several times.

Ask someone:

Is this knotted?

It may look extremely complicated.

But if you can continuously manipulate it back into a simple loop without cutting it or allowing the string to pass through itself, then mathematically it was still the unknot.

This immediately teaches an important mathematical lesson:

Complex appearance does not necessarily mean complex structure.

That principle occurs throughout mathematics.


Knot Diagrams — Turning String Into Mathematics

Working with actual string is useful, but mathematicians need a way of recording knots on paper.

They use a knot diagram.

Imagine placing the knot on a table and looking directly down at it.

Where one part of the string crosses another, we indicate which strand passes over and which passes underneath.

This distinction is crucial.

Two diagrams may contain exactly the same pattern of crossings but represent different knots if the over-and-under information changes.

This gives students an immediate practical activity.

Tie a knot.

Place it on white paper.

Photograph it from above.

Then draw its knot diagram.

Suddenly a three-dimensional object has become a two-dimensional mathematical representation.


The First Three Knots to Investigate

A very good progression is:

unknot → trefoil knot → figure-eight knot

1. The Unknot

This is the simplest possible closed loop.

Its minimum number of crossings is:

0

Of course, you can draw the unknot with lots of apparent crossings by twisting it around.

But those crossings can ultimately be removed.

That word minimum is important.


2. The Trefoil Knot

The trefoil is probably the most famous non-trivial knot.

Its simplest diagram has:

3 crossings

Try as you might, you cannot reduce it to a diagram containing zero crossings without cutting the loop or allowing one strand to pass through another.

That makes it fundamentally different from the unknot.


3. The Figure-Eight Knot

The figure-eight knot has a minimum of:

4 crossings

It is another genuinely different knot.

Now we have the beginnings of a classification system.

But we immediately encounter a mathematical difficulty.

Is counting crossings enough to identify a knot?

Unfortunately, no.


Why Counting Crossings Isn't Enough

Suppose I take a trefoil and twist part of it so that my diagram contains five crossings.

Has the trefoil suddenly become a five-crossing knot?

No.

The underlying knot has not changed.

We therefore distinguish between:

the number of crossings in a particular diagram

and

the minimum crossing number of the knot itself.

The latter is an example of a property that can help classify knots.

This leads to one of the central ideas of knot theory:

invariants.


What Is an Invariant?

An invariant is something that remains unchanged when we manipulate the knot in permitted ways.

Students already know invariants, even if they have never used the word.

For example, suppose I rearrange:

2 + 7 + 3

as:

7 + 3 + 2

The order has changed, but the total remains 12.

The total is invariant under that rearrangement.

In knot theory, mathematicians search for properties that remain unchanged when the knot is stretched and manipulated.

If two knots have different values for a suitable invariant, they cannot be equivalent.

This is a wonderfully important mathematical strategy:

Don't try to compare every detail. Find something that cannot change.


But How Do We Know When Manipulating a Diagram Is Legal?

This is where three wonderfully elegant operations enter the story.

They are called the Reidemeister moves.

They describe changes we can make to a knot diagram without changing the underlying knot.


Reidemeister Move I — The Twist

Imagine making a little loop in part of the string.

On the diagram, this introduces a crossing.

Undo the loop and the crossing disappears.

The knot itself has not changed.

This immediately explains why simply counting the visible crossings in a diagram cannot identify a knot.

You can create extra crossings just by twisting the string.


Reidemeister Move II — Two Crossings Appear or Disappear

Imagine pulling one section of string across another so that two crossings appear.

Reverse the movement and both disappear.

Again, the knot itself has not changed.

The diagram looks different.

The topology does not.


Reidemeister Move III — Sliding a Strand

The third move involves sliding one strand past a crossing formed by two other strands.

No crossings are created or destroyed.

Their arrangement simply changes.

This one can be harder to visualise on paper, so actual cord or rope makes a much better teaching tool.

And then comes a remarkable result:

Two diagrams represent the same knot if one can be transformed into the other using Reidemeister moves.

That is an extraordinarily powerful statement.


A Practical Challenge — Same Knot or Different Knot?

This can become a superb classroom or home-laboratory investigation.

Prepare several loops of differently coloured cord.

Some should be:

  • unknots disguised with twists;

  • trefoils arranged in different ways;

  • figure-eight knots;

  • mirror-image trefoils.

Label them A, B, C, D and so on.

Then ask students to classify them.

But there is an important rule.

They cannot simply say:

"They look the same."

They must provide mathematical evidence.

Can one be transformed into another?

Can crossings be removed using Reidemeister moves?

Is there an invariant that distinguishes them?

This changes the activity from puzzle solving into mathematical reasoning.


A Particularly Interesting Question — What About a Mirror?

Tie a trefoil.

Now imagine looking at it in a mirror.

Is the reflected knot the same knot?

This introduces another beautiful idea.

Some knots have chirality, or handedness.

A trefoil can be right-handed or left-handed.

The two are mirror images.

But one cannot simply be manipulated into the other using the permitted transformations.

It is rather like your hands.

Your left hand and right hand have the same basic structure.

One is the mirror image of the other.

But no matter how you rotate your left hand in ordinary three-dimensional space, you cannot turn it into a right hand.

That makes the trefoil an excellent bridge between physical intuition and abstract mathematics.


Could We Build a Knot Detective?

This is where the subject becomes even more interesting for students studying computing.

Imagine taking photographs of different knots.

Could a computer determine which knot it was looking at?

At first it sounds straightforward.

Count the crossings.

But we have already discovered the problem.

A trefoil can be drawn with more than three crossings.

An unknot can be made to look extremely complicated.

So the computer would need something more sophisticated.

It would need to recognise transformations.

It might need to calculate invariants.

It might need algorithms capable of deciding whether two knot diagrams represent the same underlying knot.

A simple piece of string has suddenly connected:

mathematics → topology → algorithms → computer science.


From GCSE Mathematics to University Mathematics

One reason I particularly like topics such as knot theory is that they challenge the idea that mathematics is simply a ladder of increasingly difficult calculations.

At GCSE, mathematics can sometimes appear to consist of:

  • solving equations;

  • calculating percentages;

  • manipulating fractions;

  • finding angles;

  • drawing graphs.

All of those are important.

But they are not the whole of mathematics.

Knot theory requires surprisingly little conventional calculation to get started.

A GCSE student can understand the central problem:

Are these two knots really the same?

Yet answering that question completely can require extremely sophisticated mathematics.

That makes it ideal for showing students what mathematics looks like beyond the examination syllabus.


Mathematics Begins With Definitions

There is another lesson hidden inside knot theory.

Mathematicians have to be extraordinarily careful about definitions.

What exactly counts as the same knot?

What transformations are allowed?

What does "crossing" mean?

What properties remain invariant?

This is not mathematical pedantry.

Without precise definitions, we cannot prove anything.

Consider two students looking at two pieces of string.

One says:

"They are obviously the same knot."

The other says:

"No they aren't."

Who is correct?

Looking is not enough.

We need a definition of equivalence and a method for establishing it.

That is mathematics.


What Does "Proof" Mean Here?

This leads to perhaps the most valuable discussion of all.

Students often encounter proof through algebra or geometry.

Prove that two angles are equal.

Prove that an expression is divisible by something.

Prove an identity.

Knot theory offers a different style of proof.

Suppose I claim that two knot diagrams represent the same knot.

One possible proof is constructive:

I show a sequence of permitted transformations that converts one into the other.

But suppose I claim they are different.

That is harder.

I cannot simply say:

"I tried for ten minutes and couldn't untangle it."

Failure to find a transformation is not proof that no transformation exists.

Instead, we need a property that distinguishes them.

That is where invariants become so powerful.

If invariant K has one value for knot A and a different value for knot B, then the knots cannot be equivalent.

This is a profound mathematical strategy:

Find something that must stay the same. Then show that it doesn't.


Going Further — Knot Polynomials

For GCSE students, I would probably stop before the algebra becomes too sophisticated.

For A-level and Further Maths students, however, we can mention that mathematicians have developed algebraic objects associated with knots.

Examples include:

  • the Alexander polynomial;

  • the Jones polynomial;

  • the HOMFLY polynomial.

The details quickly lead beyond school mathematics.

But the principle is accessible.

We take a geometric object — a knot — and associate an algebraic object with it.

Then we use algebra to investigate topology.

This is one of the recurring themes of higher mathematics:

Translate a difficult problem into another mathematical language where it becomes easier to study.


And Then There Is DNA

There is a fascinating real-world connection.

DNA molecules are extremely long compared with their width and can become twisted, looped and tangled.

Certain enzymes cut, pass and reconnect strands of DNA.

That means questions about knotting and linking become relevant when studying the geometry and topology of DNA.

Knot theory has therefore found applications in areas of molecular biology.

I would keep this secondary when teaching the topic because the mathematics is interesting enough in its own right.

But it provides a wonderful answer when a student inevitably asks:

"Does anyone actually use this?"

Yes.

What began as playing with loops of string can help provide mathematical tools for describing tangled molecules.


Try This at Home

You need very little equipment:

  • several pieces of cord or thick string;

  • tape or a method of joining the ends;

  • paper;

  • a pen;

  • ideally a camera or phone.

Start with the unknot.

Then make a trefoil.

Then make a figure-eight knot.

Photograph each from above and draw its knot diagram.

Next, deliberately make each knot look more complicated without changing the underlying knot.

Add twists.

Change its shape.

Turn it over.

Then challenge someone else to identify it.

Finally, try producing two very complicated-looking diagrams that are actually both unknots.

The lesson becomes:

Never trust the diagram alone.


A Further Maths Challenge

Here is a question worth thinking about.

Suppose two knots have the same minimum crossing number.

Does that prove they are the same knot?

No.

So crossing number is useful, but not sufficient.

This raises the next question:

What collection of invariants would be sufficient to identify every knot uniquely?

Now we are no longer asking an examination-style question.

We are beginning to think like mathematicians.


Why I Would Teach This Even Though It Isn't on the Syllabus

I think students should occasionally encounter mathematics for which there is no examination mark attached.

Knot theory is ideal.

The equipment costs almost nothing.

There is no long list of formulae to learn.

A student can begin investigating it almost immediately.

Yet underneath that simple activity lie some of the most important ideas in mathematics:

  • abstraction;

  • classification;

  • equivalence;

  • invariants;

  • transformation;

  • proof;

  • algorithms;

  • topology.

Most importantly, it challenges the idea that mathematics is primarily about calculating answers.

Sometimes mathematics asks something much deeper:

How do we know that two things are fundamentally the same?

And equally importantly:

How could we prove that they are not?


From a Piece of String to Modern Mathematics

That is what makes knot theory such a wonderful piece of mathematics beyond the syllabus.

Begin with a piece of string.

Tie a knot.

Join the ends.

Draw it.

Twist it.

Stretch it.

Try to simplify it.

Then ask one apparently innocent question:

What knot is it?

The unknot seems obvious.

The trefoil looks manageable.

The figure-eight introduces another possibility.

Then come mirror images, invariants, Reidemeister moves and knot polynomials.

Before long, the student who thought they were simply playing with string has encountered genuine topology.

And that is exactly why I enjoy taking students beyond the examination specification.

The syllabus teaches us mathematical techniques.

But occasionally stepping outside it shows us something equally important:

what mathematics actually is.

Anyone can tie a knot.

Proving which knot you have tied is considerably harder.



06 October 2026

Resonance — When a Small Push Creates a Surprisingly Large Motion

 



Resonance — When a Small Push Creates a Surprisingly Large Motion

A tiny repeated force can sometimes produce an enormous response. The secret is not how hard you push — but when you push.

Resonance is one of those physics ideas that students often meet first as a definition:

Resonance occurs when the frequency of a driving force is equal, or close, to the natural frequency of an oscillating system, producing a large-amplitude oscillation.

That definition is perfectly reasonable.

But it doesn't really convey what resonance looks like.

A much better approach is to create a mechanical oscillator, drive it through a range of frequencies and actually measure the amplitude.

At first, very little happens.

Increase the driving frequency and the oscillations become larger.

Keep going and suddenly the system seems to come alive. The mass moves backwards and forwards with a surprisingly large amplitude.

Increase the frequency further — and the amplitude falls again.

Plot amplitude against frequency and we have produced something far more useful than a dramatic demonstration.

We have measured a resonance curve.

And that simple experiment opens the door to understanding musical instruments, car suspension, machinery, bridges, buildings and even why engineers sometimes need to design structures specifically to avoid resonance.


Everything That Oscillates Has Its Own Preferences

Imagine hanging a mass from a spring.

Pull it down slightly and release it.

It oscillates.

Ignoring damping for the moment, the system has a characteristic frequency at which it naturally wants to oscillate. We call this its natural frequency.

For an ideal mass-spring system:

f = (1 / 2pi) sqrt(k / m)

where:

  • f is the natural frequency in Hz;

  • k is the spring constant in N/m;

  • m is the oscillating mass in kg.

This immediately gives us some useful predictions.

Increase the mass and the natural frequency decreases.

Use a stiffer spring and the natural frequency increases.

That alone makes an excellent investigation.

But resonance appears when we introduce something else.

Instead of simply pulling the mass down and releasing it, we continually apply an external periodic force.

We drive the oscillator.


The Importance of Timing

A playground swing provides perhaps the most familiar example.

Imagine pushing somebody on a swing.

If you push at random moments, your pushes may not help very much. Some might even oppose the motion.

But push at approximately the right point during every oscillation and something very different happens.

Each push adds energy.

The swing goes higher.

Another correctly timed push adds more energy.

It goes higher again.

The individual pushes do not need to be enormous.

What matters is their timing.

That is the essential idea behind resonance.


Turning Resonance Into an Experiment

There are many ways of demonstrating mechanical resonance.

A spring-mass arrangement is particularly useful because it allows the effect to become quantitative.

We need an oscillator with a natural frequency and some means of applying a periodic driving force whose frequency can be varied.

We then investigate what happens as the driving frequency is gradually changed.

For example, we might test:

0.5 Hz

0.7 Hz

0.9 Hz

1.1 Hz

1.3 Hz

1.5 Hz

and so on.

At every frequency, allow the system to settle and measure its oscillation amplitude.

The important thing is to change frequency, while keeping other important variables as consistent as reasonably possible.

We can then construct a table:

Driving frequencyOscillation amplitude
LowSmall
IncreasingIncreasing
Near resonanceMaximum
Above resonanceDecreasing

The actual numerical measurements can then be plotted.

Horizontal axis: driving frequency

Vertical axis: amplitude

Now the experiment becomes particularly interesting.


The Resonance Curve Appears

The resulting graph should not simply continue upwards.

Instead, it rises towards a peak and then falls.

That peak tells us something important about the oscillator.

It identifies the region of its resonant frequency.


This is a much richer experiment than simply watching a spring shake violently.

We can ask:

At what frequency was the greatest amplitude observed?

How does that compare with the calculated natural frequency?

How broad is the resonance peak?

What happens if we increase the damping?

What happens if we change the mass?

What happens if we use a different spring?

Suddenly a simple demonstration has become an experimental investigation.


Natural Frequency and Resonant Frequency

At school level we often say that resonance occurs when:

driving frequency = natural frequency

That is an extremely useful model and captures the central physics.

In a real damped oscillator, however, the precise frequency at which maximum displacement occurs can be slightly different from the undamped natural frequency.

For most introductory experiments the important observation is therefore that maximum response occurs near the natural frequency.

That distinction becomes increasingly important as students progress further into oscillations.


Why Doesn't the Amplitude Become Infinite?

There is an interesting question hiding inside the experiment.

If every correctly timed push adds more energy, why doesn't the amplitude simply keep increasing forever?

Because real oscillators lose energy.

There is always some form of damping.

Energy may be dissipated through:

  • air resistance;

  • friction;

  • deformation of materials;

  • friction within bearings or supports;

  • internal losses in the spring.

Eventually the energy being supplied by the driving force balances the energy being dissipated.

A steady amplitude is reached.

Damping is therefore not merely an irritating experimental complication.

It is an important part of the physics.


Now Add Damping

This gives us an excellent second experiment.

Repeat the resonance investigation but increase the damping.

Depending on the apparatus, this might be achieved using a vane moving through air or liquid, magnetic damping or another suitable controlled mechanism.

Construct another resonance curve.

What changes?

The resonance peak becomes lower and broader.

Greater damping means that energy is removed from the oscillator more rapidly, preventing such a large amplitude developing.

This leads to a fascinating engineering question:

Do we want resonance or don't we?

The answer depends entirely on what we are designing.


Sometimes Resonance Is Exactly What We Want

Resonance is not inherently dangerous.

In many technologies it is extraordinarily useful.

Musical instruments

Musical instruments depend heavily upon resonant behaviour.

A vibrating string by itself moves very little air.

Connect it to the body of a guitar, violin or piano and resonant structures help transfer energy into the surrounding air.

Wind instruments contain resonating columns of air.

Organ pipes are wonderful examples.

The dimensions of the pipe help determine the frequencies at which strong resonances occur.

Music itself provides an excellent reminder that resonance is not some unusual laboratory phenomenon.

It surrounds us.

Radio and electronics

Electrical circuits can also resonate.

A tuned circuit can respond strongly to a particular range of frequencies while responding much less strongly to others.

This principle is fundamental to selecting signals.

The mechanical resonance experiment therefore introduces an idea that reappears throughout physics.


Sometimes Resonance Is Exactly What We Don't Want

Now turn the problem around.

Suppose the oscillator isn't a musical instrument.

Suppose it is a bridge.

Or a building.

Or part of a machine.

Repeated forces may occur because of:

  • rotating machinery;

  • engines;

  • motors;

  • road traffic;

  • wind;

  • waves;

  • earthquakes;

  • repeated human movement.

If one of those driving frequencies approaches a natural frequency of the structure, the response can become much larger than might otherwise be expected.

Engineers therefore need to understand both natural frequencies and damping.

The objective is often not simply to make something "strong".

It is to control how it responds dynamically.


The Famous Bridge Example — But With an Important Correction

Resonance discussions frequently mention the collapse of the Tacoma Narrows Bridge in 1940.

It certainly provides a spectacular example of oscillating structures and the importance of understanding dynamic behaviour.

But describing the final collapse simply as ordinary forced resonance is misleading.

The bridge's dramatic motion involved aeroelastic effects, particularly torsional flutter, in which the interaction between the bridge and airflow sustained and amplified the oscillations.

That distinction matters.

One of the things I particularly like about practical physics is that experiments encourage us to move beyond simplified slogans.

The spring-mass experiment teaches genuine principles that engineers use — but real structures can involve several interacting effects simultaneously.


Buildings Have Natural Frequencies Too

A tall building may appear completely stationary.

It isn't.

Structures can bend and oscillate.

Engineers therefore consider how buildings respond to wind and earthquakes.

One fascinating solution is the tuned mass damper.

Imagine a very large mass deliberately installed within a building.

It is designed so that its motion helps oppose unwanted movement of the structure and dissipate energy.

The same fundamental physics that we investigate with a mass and spring in a laboratory can therefore appear on an architectural scale.

That is one of the things I find so appealing about teaching physics experimentally.

The apparatus might fit on a bench.

The physics may apply to something hundreds of metres tall.


Your Car Is an Oscillator

Push down firmly on the corner of a stationary car and release it.

The suspension responds.

The combination of vehicle mass, springs, tyres and dampers forms an oscillating system.

Without adequate damping, the car could continue bouncing after every bump.

Too much or badly chosen damping, however, would produce different problems.

Vehicle suspension therefore provides another excellent real-world application of the same ideas:

mass + elasticity + damping + external forcing.

A rough road continually supplies disturbances.

The suspension has to control the response.


A Particularly Good Student Investigation

For an A-level student, I would develop the experiment in stages rather than trying to investigate everything simultaneously.

Investigation 1 — Find the natural frequency

Displace the mass slightly and release it.

Rather than timing one oscillation, measure the time for perhaps ten complete oscillations.

Then:

T = total time / number of oscillations

and:

f = 1 / T

This reduces the percentage uncertainty compared with timing a single oscillation.


Investigation 2 — Predict the natural frequency

Measure or determine the spring constant k and the oscillating mass m.

Calculate:

f = (1 / 2pi) sqrt(k / m)

Compare the theoretical result with the experimental measurement.

Already there is plenty to discuss:

  • uncertainty;

  • effective mass of the spring;

  • damping;

  • assumptions in the mathematical model;

  • measurement technique.


Investigation 3 — Produce the resonance curve

Apply a periodic driving force.

Gradually vary its frequency.

At each frequency, allow transient behaviour to settle before recording the steady-state amplitude.

Plot:

amplitude against driving frequency

Look for the peak.

Does its position agree reasonably well with the natural frequency measured earlier?

Now we have connected theory, free oscillation and forced oscillation.


Investigation 4 — Change the damping

Repeat selected measurements with additional damping.

Plot both sets of results on comparable axes.

Students should now be able to see rather than merely memorise that greater damping produces a lower, broader resonance response.


Modern Sensors Can Make This Even Better

This is also an excellent experiment for electronic data collection.

A motion sensor, position sensor, accelerometer or suitable force sensor can collect far more information than we could obtain simply by watching the mass.

A position-time graph can reveal the oscillation directly.

Digital data can also make it easier to determine:

  • frequency;

  • amplitude;

  • period;

  • phase relationships;

  • transient behaviour;

  • steady-state behaviour.

With suitable data-logging equipment, the experiment can become an excellent bridge between traditional practical physics and modern experimental analysis.

It is particularly satisfying when students can see the oscillator moving in front of them while its motion is simultaneously appearing as a graph.

The graph stops being an abstract mathematical object.

It becomes a record of something they have actually watched happen.


Don't Rush Through the Resonant Frequency

There is an experimental trap here.

Suppose measurements are taken at:

1 Hz, 2 Hz, 3 Hz, 4 Hz, 5 Hz...

and resonance happens near 2.4 Hz.

The experiment may almost completely miss the interesting part of the curve.

A much better strategy is to begin with relatively large frequency intervals.

Once the approximate position of resonance has been identified, take many more measurements around that region.

For example:

2.0 Hz

2.1 Hz

2.2 Hz

2.3 Hz

2.4 Hz

2.5 Hz

2.6 Hz

This itself teaches an important experimental skill.

Good experimental design does not always mean taking equally spaced measurements regardless of what the data are telling you.

Sometimes the preliminary results should determine where you investigate in greater detail.


Watch the Phase as Well as the Amplitude

For more advanced students, there is another layer waiting to be discovered.

Amplitude is not the only thing changing as we alter the driving frequency.

The phase relationship between the driving force and the oscillator also changes.

At low frequencies, the oscillator responds relatively closely to the driver.

Around resonance, the phase relationship changes significantly.

At frequencies well above resonance, the displacement response becomes increasingly out of phase with the driving force.

This is where resonance becomes much more than "the biggest wobble".

It becomes part of the wider physics of driven harmonic motion.


A Simple Experiment With Enormous Reach

This is exactly the sort of practical physics I enjoy.

On the bench we have something extremely simple:

a mass,

a spring,

a driver,

and a measuring system.

Yet from that apparatus we can discuss:

  • simple harmonic motion;

  • natural frequency;

  • forced oscillations;

  • energy transfer;

  • damping;

  • resonance curves;

  • phase;

  • musical instruments;

  • vehicle suspension;

  • buildings;

  • bridges;

  • machinery;

  • electronic circuits.

That is an extraordinary amount of physics from one oscillating mass.


The Most Important Lesson Isn't "Resonance Makes Things Shake"

The most useful lesson is more subtle.

A system's response depends not only on how large the applied force is, but also on how that force varies with time.

A comparatively small periodic force, applied at the wrong frequency, may produce little response.

The same force applied near a natural frequency may produce a dramatically larger response.

That is why understanding natural frequencies matters so much in engineering.

And it is why simply showing students an object vibrating violently misses an opportunity.

Measure it.

Change the frequency.

Plot the amplitude.

Find the peak.

Change the damping.

Then ask why the graph changed.

At that point resonance stops being a definition to remember for an examination.

It becomes something students have discovered for themselves.


Final Thought

Perhaps the most remarkable thing about resonance is that the driving force does not necessarily need to be enormous.

It simply needs to keep supplying energy at the right time.

Sometimes in physics, timing matters more than force.

And one small mass bouncing on a spring can reveal why.

What Actually Happens When a Cold Front Arrives?

  What Actually Happens When a Cold Front Arrives? Look at almost any television weather forecast and sooner or later you will see a line mo...