Approximating Values of a Function Using Local Linearity and Linearization

students, imagine you are looking at a graph of a hill and need a quick estimate of how high it is at a point just a tiny step away from where you already know the height. You could calculate the exact value, but sometimes the exact method is slow or impossible by hand. Calculus gives a powerful shortcut: near a point, many functions behave almost like straight lines 📈. This idea is called local linearity, and the formula built from it is called a linearization.

In this lesson, you will learn how to use the tangent line to approximate function values, why the method works, and when it is most accurate. By the end, you should be able to explain the idea in words, write the approximation formula, and apply it to real situations like estimating temperatures, distances, or small changes in measurements.

Why Local Linearity Works

A smooth function usually looks curved when you zoom out, but if you zoom in close enough near a point $x=a$, the graph often looks almost like its tangent line. That is the big idea behind local linearity. The tangent line gives the best straight-line approximation to the function near that point.

If $f$ is differentiable at $x=a$, then the graph of $f$ is well-approximated near $a$ by the line

$$L(x)=f(a)+f'(a)(x-a).$$

This equation is called the linearization of $f$ at $x=a$. It uses two pieces of information:

$f(a)$, the exact function value at the base point
$f'(a)$, the slope of the tangent line at that point

The meaning is simple: start at the known point $(a,f(a))$, then move horizontally from $a$ to a nearby $x$ value, and use the slope to estimate the change in $f(x)$.

For example, if $f(x)=\sqrt{x}$ and you want to estimate $\sqrt{10}$, you may choose $a=9$ because $\sqrt{9}=3$ is easy to compute. Since $f'(x)=\frac{1}{2\sqrt{x}}$, we have $f'(9)=\frac{1}{6}$. Then

$$L(x)=3+\frac{1}{6}(x-9).$$

$$\sqrt{10}\approx L(10)=3+\frac{1}{6}=\frac{19}{6}.$$

That estimate is close to the actual value because $10$ is near $9$.

Building the Linearization Step by Step

To use local linearity on AP Calculus BC, follow a reliable process ✅:

Choose a base point $a$ where $f(a)$ and $f'(a)$ are easy to find.
Write the linearization formula $L(x)=f(a)+f'(a)(x-a)$.
Substitute the target value of $x$ into $L(x)$.
Interpret the result as an approximation, not an exact value.

A classic example is estimating $\sin(0.1)$.

Choose $a=0$ because $\sin(0)=0$ and $\cos(0)=1$. Let $f(x)=\sin(x)$. Then $f'(x)=\cos(x)$, so

$$L(x)=0+1(x-0)=x.$$

Thus

$$\sin(0.1)\approx L(0.1)=0.1.$$

The actual value is slightly less than $0.1$, but the approximation is very good because $0.1$ is close to $0$.

Notice the connection to tangent lines: the linearization is not a new object separate from the tangent line. It is the tangent line written as a function so that it can be used for approximation.

Interpreting Error and Accuracy

The closer $x$ is to $a$, the better the linear approximation usually is. That is because the function and its tangent line match exactly at $x=a$ and have the same slope there. However, as you move farther away, the curve may bend away from the line.

The sign of the error depends on the shape of the graph:

If the graph is concave up, the tangent line often lies below the curve near the point.
If the graph is concave down, the tangent line often lies above the curve near the point.

For example, $f(x)=e^x$ is concave up everywhere. If you linearize at $a=0$, then $f(0)=1$ and $f'(0)=1$, so

$$L(x)=1+x.$$

Using this, we estimate

$$e^{0.2}\approx 1.2.$$

The actual value is about $1.221$, so the estimate is a little low, which matches the fact that $e^x$ is concave up.

A useful AP idea is that local linearity is strongest when the input change is small. If the problem asks for a value far from the base point, the estimate may be less reliable. In that case, choosing a different $a$ closer to the target can improve accuracy.

Real-World Meaning of Linear Approximation

Local linearity is useful whenever exact computation is difficult but a nearby value is known. This happens in science, engineering, and everyday measurement 🔧.

Suppose a circular metal plate has radius $r$ and area $A=\pi r^2$. If the radius changes only a tiny amount, the area changes in a way that is almost linear near the current value of $r$.

If $r=10$ cm and you want to estimate the area when $r=10.1$ cm, let

$$f(r)=\pi r^2.$$

Then

$$f'(r)=2\pi r,$$

so at $r=10$,

$$f(10)=100\pi, \quad f'(10)=20\pi.$$

The linearization is

$$L(r)=100\pi+20\pi(r-10).$$

Now substitute $r=10.1$:

$$L(10.1)=100\pi+20\pi(0.1)=102\pi.$$

This gives a quick estimate of the new area without squaring $10.1$ directly.

Another practical example is estimating how a function changes in response to a small measurement error. If a thermometer reads a temperature near a value where a formula is known, local linearity can estimate the effect of a tiny change. Calculus often turns complicated nonlinear relationships into manageable straight-line approximations.

How to Write and Use the Approximation Formally

On the AP exam, it is important to use correct notation and language. If $f$ is differentiable at $x=a$, then the linear approximation near $a$ is

$$f(x)\approx f(a)+f'(a)(x-a).$$

The symbol $\approx$ means “approximately equal to,” not exactly equal to. That distinction matters.

A good written response should include:

the chosen function $f$
the base point $a$
the derivative $f'(x)$ or the slope at $a$
the linearization $L(x)$
the final approximation with units if the situation has units

For example, if a problem asks for an approximation of $\ln(1.05)$, choose $f(x)=\ln(x)$ and $a=1$ because $\ln(1)=0$ and $f'(x)=\frac{1}{x}$, so $f'(1)=1$.

Then

$$L(x)=0+1(x-1)=x-1,$$

and

$$\ln(1.05)\approx L(1.05)=0.05.$$

This is a standard AP-style estimate because $1.05$ is close to $1$.

Common Mistakes to Avoid

students, students often lose points on linearization problems for a few common reasons:

Using the wrong base point $a$
Forgetting to compute $f'(a)$ correctly
Writing the tangent line but not using it to estimate the requested value
Treating an approximation as an exact value
Ignoring units in applied problems

Another mistake is choosing a base point that is not convenient. The best base point is usually one where both $f(a)$ and $f'(a)$ are simple numbers. For example, $a=0$, $a=1$, $a=4$, or $a=9$ are often useful for functions involving $e^x$, $\ln(x)$, $\sqrt{x}$, or powers.

It also helps to remember that linearization is a local tool. If the target value is too far from the base point, a calculator or another method may give a better estimate. Still, AP Calculus expects you to recognize when linearity is appropriate and how to justify the result.

Connecting Linearization to the Bigger Picture

Local linearity is part of the broader AP Calculus BC topic of contextual applications of differentiation because it uses derivative information to understand real behavior. Derivatives describe rates of change, and linearization uses that rate to predict nearby values.

This same idea also appears in many other calculus topics:

In motion, derivatives estimate velocity and acceleration.
In related rates, derivatives connect changing quantities.
In optimization, derivatives help identify maximum and minimum values.
In approximations, derivatives produce quick estimates.

So when you use linearization, you are doing more than just estimating a number. You are using derivative information to model how a function behaves in the neighborhood of a point. That is a central theme of calculus.

Conclusion

Local linearity says that smooth functions look almost like straight lines when you zoom in close enough. The linearization formula

$$L(x)=f(a)+f'(a)(x-a)$$

gives a powerful way to approximate function values near $x=a$. It works best when the target input is close to the base point and when the function is differentiable there. On AP Calculus BC, you should be able to choose a convenient base point, compute the derivative, write the tangent-line approximation, and interpret the result carefully. This skill is valuable because it turns difficult calculations into quick and meaningful estimates 📚.

Study Notes

Local linearity means a differentiable function behaves like its tangent line near a point.
The linearization of $f$ at $x=a$ is $L(x)=f(a)+f'(a)(x-a)$.
Use a base point where $f(a)$ and $f'(a)$ are easy to find.
Linearization gives an approximation, so use $\approx$, not $=$.
The closer the input is to $a$, the more accurate the estimate usually is.
Concavity affects whether the tangent line lies above or below the curve near the base point.
Linearization is useful in real situations involving small changes in measurements, geometry, and science.
On AP Calculus BC, show the function, derivative, linearization, and final estimate clearly.