AP® CALCULUS AB & BC · UNIT 4

Local Linearity and Linearization

Use tangent lines to estimate nearby function values. Work through square roots, exponentials, and a temperature estimate, then use concavity to tell whether the estimate is high or low.

Topic 4.6 study notes3 worked examples · 3 practice problemsUpdated 10 November 2026
Calculus notebook with limit and integral symbols, a compass and pencil on an olive and cream study desk

1. Use a tangent line to estimate

Near a point where f is differentiable, its graph closely follows the tangent line. This is local linearity. The linearization L gives an estimate at a nearby input x.

L(x)=f(a)+f′(a)(x−a)
f(x)≈L(x)

The tangent line starts at the known height f(a). Its slope f′(a) times the input change x−a gives the estimated change in the output. At the anchor a, the curve and line have the same height and slope.

In Taylor notation, L is the first-order Taylor polynomial centered at a.

2. Choose an anchor and check concavity

  1. Choose an anchor.

    Find a nearby input in the domain where you can evaluate the function and its derivative.

  2. Build the tangent line.

    Use f(a) as the starting value and f′(a) as the slope.

  3. Evaluate at the target.

    Substitute the requested input into L. State the result as an approximation in the original output units.

  4. Check the direction of the error.

    Use the concavity between the anchor and the target to compare the tangent line with the curve.

For x≠a, use the table when the second derivative keeps the stated strict sign between the anchor and target.

Concavity and the tangent estimate
Curve shapeTangent estimate
Concave up
f″(u)>0
Underestimate
L(x)<f(x)
Concave down
f″(u)<0
Overestimate
L(x)>f(x)

Read the whole interval. An increasing slope places the tangent below the curve. A decreasing slope places it above. At x=a, the estimate is exact.

3. Three worked examples

Worked example 1

Estimate a square root near a perfect square

Use local linearity to estimate 25.2.

1 · Choose the known square root

Let f(x)=x and choose a=25. Then f(25)=5.

2 · Find the slope and write the tangent line

f′(x)=12x
f′(25)=1225=0.1
L(x)=5+0.1(x−25)

3 · Estimate the value and compare with the curve

Evaluate the tangent line at x=25.2.

L(25.2)=5+0.1(25.2−25)
L(25.2)=5+0.1(0.2)=5.02
25.2≈5.02

For x>0, f″(x)=−14xx<0. The curve is concave down from 25 to 25.2, so 5.02 is an overestimate.

Worked example 2

Estimate an exponential near zero

Use a tangent line to estimate e0.03.

1 · Choose the anchor and slope

For f(x)=ex, choose a=0. The value is f(0)=1. Since f′(x)=ex, the slope is f′(0)=1.

2 · Build and evaluate the linearization

L(x)=1+1(x−0)=1+x
L(0.03)=1+0.03=1.03

3 · Interpret the estimate

e0.03≈1.03

Since f″(x)=ex>0 throughout the interval, 1.03 is an underestimate.

Worked example 3

Estimate a cooling temperature from its rate

At t=10 minutes, a water bath has temperature T(10)=68°C and rate T′(10)=−0.8°Cmin. The temperature function is concave up between 10 and 10.5 minutes. Estimate T(10.5) and classify your estimate.

1 · Use the measured temperature and rate

L(t)=68−0.8(t−10)

The anchor rate is negative, so the tangent predicts a drop in temperature.

2 · Evaluate half a minute later

L(10.5)=68−0.8(0.5)=67.6

3 · State the units and estimate type

The estimated temperature is 67.6°C. The concave-up curve lies above its tangent, so the estimate is low.

4. Read overestimates and underestimates from graphs

The solid curve is the function. The dashed line is its linearization. Each vertical guide compares the two heights at the same input, using a point far enough from the anchor for the gap to be visible.

Square root and its tangent at 25

9 16 25 36 Input x 3 4 5 6 Value A
The curve and tangent meet at (25,5). At the comparison input 9, the square on the tangent is at 3.4, above the curve value 3. The same concavity makes 5.02 an overestimate in Example 1.

Exponential and its tangent at zero

−1 0 1 Input x 0 1 2 3 Value A
The curve and tangent meet at (0,1). At the comparison input 1, the square on the tangent is at 2, below the curve value e. The same concavity makes 1.03 an underestimate in Example 2.

5. Three practice problems

Try each problem before opening its notebook solution.

Practice 1. Estimate a sine value near zero

Approximate sin(0.12) using the linearization of f(x)=sin(x) at x=0. Use radians.

Show solution 1

Notebook solution

1 · Find the anchor height and slope

At a=0, f(0)=0 and f′(0)=cos(0)=1.

L(x)=0+1(x−0)=x

2 · Evaluate and check concavity

L(0.12)=0.12

Thus sin(0.12)≈0.12. Since f″(x)=−sinx<0 for 0<x<0.12, this is an overestimate.

Practice 2. Estimate a natural logarithm near one

Estimate ln(1.05) using the tangent line to f(x)=ln(x) at x=1.

Show solution 2

Notebook solution

1 · Evaluate the function and derivative at one

At a=1, f(1)=0 and f′(1)=11=1.

L(x)=0+1(x−1)=x−1

2 · Evaluate and classify the estimate

L(1.05)=0.05

Thus ln(1.05)≈0.05. The logarithm is concave down for x>0, so this is an overestimate.

Practice 3. Estimate a square root near sixteen

Use the linearization of f(x)=x at x=16 to estimate 16.5.

Show solution 3

Notebook solution

1 · Find the tangent line at sixteen

At a=16, f(16)=4.

f′(16)=1216=18
L(x)=4+18(x−16)

2 · Evaluate at the target input

L(16.5)=4+18(0.5)
L(16.5)=4+0.0625=4.0625

Thus 16.5≈4.0625. Concavity down on the interval makes this an overestimate.

6. Check an AP response

  • Identify the anchor. Give a, f(a), and f′(a). These establish the point of tangency and slope.
  • Show the line and the evaluation. Write L(x), substitute the target, and use ≈ when comparing with f(x).
  • Justify the estimate type. Name the interval and state how the concavity places the tangent above or below the curve.
  • Include the units in context. The estimate has the same units as the function value.

7. Continue studying Unit 4

Review the meaning of a derivative or continue to indeterminate limits.