Differential Equations: Calculus That Models Change

An equation whose answer is a function

In ordinary algebra, an equation like x^2 = 9 asks you to find a number — here x = 3 or x = -3. A differential equation asks something deeper: it relates a function to its own derivatives, and the answer you are hunting for is an entire function, not a single value. You are solving for the whole story of how a quantity behaves over time, given only a rule about how it changes.

Why is this the natural language of science? Because the things we can actually measure in the world are usually rates. A thermometer that is cooling, a population that is breeding, money that is earning interest — in each case the easy thing to state is the rate of change right now, and that rate often depends on the current amount. Writing 'the rate equals (some rule about the current state)' is exactly writing a differential equation.

The most famous one: growth and decay

The cleanest differential equation in all of science is dy/dt = k*y. In words: the rate at which y changes is proportional to how much y there already is. When a quantity's growth feeds on its own size, this is the rule. Bigger bank balance earns more interest; a larger colony of bacteria makes more bacteria per hour; more radioactive atoms means more decays per second.

What function has a derivative proportional to itself? The exponential. The solution is y(t) = y0 * e^(k*t), where y0 is the starting amount at t = 0. If k > 0 you get runaway exponential growth; if k < 0 you get smooth exponential decay toward zero. Notice the answer is genuinely a function of t — feed it any time and it tells you the amount, the full trajectory predicted from a single rule about rate.

the rule (the equation):    dy/dt = k * y
the solution (a function):  y(t)  = y0 * e^(k*t)        y0 = y(0)

check: differentiate y(t)  ->  k * y0 * e^(k*t) = k * y(t).  it fits.

  k > 0  ->  exponential growth
  k < 0  ->  exponential decay

Newton's cooling, and seeing solutions without solving

Pure exponential growth is rare in the real world because nothing grows forever. A more lifelike example is Newton's law of cooling: a hot cup of coffee loses heat fast when it is far above room temperature and slowly when it is nearly there. Letting T be the coffee's temperature and R the room's, the rule is dT/dt = -k*(T - R) — the rate of cooling is proportional to the gap between the object and its surroundings.

Here is a beautiful idea: you can often see how solutions behave before solving anything, using a slope field. At each point (t, y) on a grid, the equation tells you the slope dy/dt right there, so you draw a tiny dash with that slope. The dashes form a current, like iron filings around a magnet. A solution is just any curve that flows along the dashes — and its starting point decides which curve you ride.

1. Read the cooling rule dT/dt = -k*(T - R) as a sentence: 'how fast the temperature drops depends on how far it is above the room.'
2. Check the extremes. When T is far above R, the gap is large, so the slope is steeply negative — fast cooling. When T nears R, the gap shrinks toward 0, so the slope flattens — cooling crawls to a stop.
3. Read off the destination. The slope is 0 exactly when T = R, so every solution drifts toward room temperature and levels off there. You learned the long-run behavior without ever finding a formula.
4. If you do want the formula, it is T(t) = R + (T0 - R)*e^(-k*t): the gap above the room decays exponentially — exactly the picture the slope field already showed you.

Where this opens onto the real world

Once you can write 'rate = rule', differential equations appear everywhere. Physics is built on them: Newton's F = m*a is secretly the equation m*(d^2 x/dt^2) = F, a relation involving the second derivative (acceleration), whose solution is the object's whole path through time. A pendulum, a planet's orbit, a vibrating string, the flow of heat — all are differential equations waiting to be read.

Beyond physics, epidemiology models an epidemic by tracking how fast the number of infected people rises as a function of who is currently infected and susceptible — that is the family of SIR equations that guided real pandemic policy. Finance prices options with the Black-Scholes differential equation. Ecology balances predators and prey with coupled equations. The same small idea — describe the change, then recover the function — quietly runs much of quantitative science.