6.2 Approximating Areas with Riemann Sums

Riemann sums estimate the area under a curve, or a definite integral, by adding the areas of rectangles or trapezoids. The four main types are left, right, midpoint, and trapezoidal sums, and you can build them from a graph, a table, an equation, or a verbal description. For AP Calculus, include units when the values come from a real context.

Riemann Sums Summary

For AP Calculus Topic 6.2, Riemann sums approximate a definite integral by breaking an interval into subintervals and adding simple areas. Left, right, and midpoint sums use rectangles; trapezoidal sums use trapezoids. The setup works from tables, graphs, formulas, or verbal descriptions.

The exam often tests two things at once: computing the approximation and explaining its error. Increasing/decreasing behavior decides whether left and right sums overestimate or underestimate, while concavity decides the direction of error for midpoint and trapezoidal sums.

Why This Matters for the AP Calculus Exam

Approximating definite integrals shows up across AP Calculus because real data often comes as a table of values, not a clean formula. You will be asked to set up and compute left, right, midpoint, and trapezoidal sums, then explain whether the result is an overestimate or underestimate. This skill connects directly to the idea that a definite integral is the limit of these sums as the rectangles get thinner, which is the foundation for everything that follows in integration and accumulation.

Approximation also matters because many functions do not have an easy antiderivative. When you cannot integrate exactly, a numerical approach (with or without a calculator) is how you get an answer. Being precise with units and setup is important for clear exam work, especially when a problem gives you data in context.

Key Takeaways

• A Riemann sum approximates the area under a curve by adding up areas of rectangles (left, right, midpoint) or trapezoids.
• Find the width of each subinterval with (b - a)/n for a uniform partition; partitions can also be nonuniform with different widths.
• Left sums underestimate increasing functions and overestimate decreasing functions; right sums do the opposite.
• For trapezoidal and midpoint sums, concavity decides the error: trapezoidal overestimates when concave up, midpoint underestimates when concave up.
• You can build any of these sums from a graph, a table of values, an equation, or a verbal description.
• As the number of subintervals grows and widths shrink toward 0, the Riemann sum approaches the exact value of the definite integral.

The Four Riemann Sums

As you move into integral calculus, the goal shifts from finding an instantaneous rate of change to finding the area under a curve. That area is hard to get exactly with geometry alone, so you approximate it with simple shapes. The more shapes you use, the closer your approximation gets to the true value.

There are four main types:

1. Left Riemann sum
2. Right Riemann sum
3. Midpoint Riemann sum
4. Trapezoidal sum

Left and Right Riemann Sums

Left and right refer to which point on each subinterval sets the height of the rectangle. A left Riemann sum uses the left endpoint of each subinterval for the height; a right Riemann sum uses the right endpoint.

Subintervals (Partitions)

You also choose how wide each base is. A uniform partition uses equal widths; a nonuniform partition uses different widths. For a uniform partition over $[a, b]$ with $n$ subintervals, each width is:

$$\Delta x = \frac{b - a}{n}$$

The pieces you divide the interval into are called subintervals.

Overestimate vs Underestimate for Left and Right Sums

Whether a left or right sum is too high or too low depends on whether the function is increasing or decreasing:

• A left Riemann sum underestimates when a function is increasing.
• A right Riemann sum overestimates when a function is increasing.
• A left Riemann sum overestimates when a function is decreasing.
• A right Riemann sum underestimates when a function is decreasing.

A quick way to remember it: for an increasing function, the right endpoint is the tallest point in each strip, so right sums run high and left sums run low. Flip it for a decreasing function.

Trapezoidal and Midpoint Sums

Both of these aim to reduce error compared to plain left or right sums.

Trapezoidal sum: instead of flat-topped rectangles, you connect each pair of points with a straight line, forming trapezoids. The area of one trapezoid is:

$$\frac{a + b}{2} \cdot h$$

where $a$ and $b$ are the two parallel side heights and $h$ is the width. The trapezoidal sum equals the average of the left and right sums, so $T_n = \frac{L_n + R_n}{2}$. It is written $T_n$.

Midpoint sum: use the height of the function at the middle of each subinterval as the rectangle's height. It is written $M_n$.

Overestimate vs Underestimate for Trapezoidal and Midpoint Sums

Here concavity is what matters, not increasing/decreasing:

• A trapezoidal Riemann sum overestimates when a function is concave up.
• A trapezoidal Riemann sum underestimates when a function is concave down.
• A midpoint Riemann sum underestimates when a function is concave up.
• A midpoint Riemann sum overestimates when a function is concave down.

Because trapezoidal and midpoint sums err in opposite directions, the true value often sits between them.

Worked Example: Building Sums from an Equation

Given $f(x) = \frac{1}{2}x^2$, find $L_4$, $R_4$, $T_4$, and $M_4$ over the interval $[0, 4]$.

Start by finding the width of each subinterval:

$$\frac{4 - 0}{4} = 1$$

Build a table of values at the endpoints:

Left sum uses $x = 0, 1, 2, 3$. Since the width is 1, sum the heights:

$$L_4 = 0 + 0.5 + 2 + 4.5 = \boxed{7}$$

Right sum uses $x = 1, 2, 3, 4$:

$$R_4 = 0.5 + 2 + 4.5 + 8 = \boxed{15}$$

Trapezoidal sum averages each pair of adjacent heights times the width:

$$\frac{0+0.5}{2}\cdot 1+\frac{0.5+2}{2}\cdot 1+\frac{2+4.5}{2}\cdot 1+\frac{4.5+8}{2}\cdot 1=\boxed{11}$$

Notice this matches $\frac{L_4 + R_4}{2} = \frac{7 + 15}{2} = 11$.

Midpoint sum uses the heights at $x = 0.5, 1.5, 2.5, 3.5$:

$$M_4 = 0.125+1.125+3.125+6.125=\boxed{10.5}$$

Are $T_4$ and $M_4$ over or underestimates? This function is concave up, so $T_4$ is an overestimate and $M_4$ is an underestimate. The true area lies between 10.5 and 11.

How to Use This on the AP Calculus Exam

Free Response

Many free-response problems give you a table of values for a rate function and ask for an approximation. Write out the full setup, not just the final number. For a trapezoidal estimate with unequal widths, keep each term separate so your work is clear. Always include units when the problem is in context, since the unit of the area is the rate's unit times the independent variable's unit.

MCQ

Multiple-choice questions often test the over/underestimate rules directly. Read whether the function is increasing/decreasing (for left and right sums) or concave up/down (for trapezoidal and midpoint sums) and apply the matching rule. Watch for nonuniform partitions where each subinterval has a different width.

Problem Solving

When data comes from a table, you do not need a formula. Match each subinterval to its width and the correct sample point (left endpoint, right endpoint, or midpoint). Factor out the width when it is constant to speed up the arithmetic, like $L_6 = \frac{1}{3}(0 + \frac{10}{9} + \frac{22}{9} + 4 + \frac{52}{9} + \frac{70}{9})$.

Common Trap

A midpoint sum needs the function value at the middle x of each subinterval, not the average of the endpoint heights. The average-of-endpoints idea belongs to the trapezoidal sum.

Practice Problems

1. Given $f(x) = 2x + 1$ over $[1, 5]$, calculate the right Riemann sum with $n = 4$.
2. Given $f(x) = x^2 + 3x$ over $[0, 2]$, compute the left Riemann sum with $n = 6$.
3. Given $f(x) = 3x^2 - 4$ over $[2, 8]$, find $T_{12}$ and $M_{12}$.

Solutions

Problem 1: Width is $\frac{5 - 1}{4} = 1$. Right endpoints are $x = 2, 3, 4, 5$.

$$(5\cdot1)+(7\cdot 1)+(9\cdot1)+(11\cdot1)=\boxed{32}$$

Problem 2: Width is $\frac{2 - 0}{6} = \frac{1}{3}$. Left endpoints start at $x = 0$.

Factor out the width:

$$L_6=\frac{1}{3}\Bigg(0+\frac{10}{9}+\frac{22}{9}+4+\frac{52}{9}+\frac{70}{9}\Bigg)\approx\boxed{7.04}$$

Problem 3: Width is $\frac{8 - 2}{12} = \frac{1}{2}$.

For $T_{12}$, use heights at every endpoint:

Using the trapezoidal pattern (each interior height counts twice, endpoints once), this works out to:

$$0.5(8+14.75+23+32.75+44+56.75+71+86.75+104+122.75+143+164.75+188)=\boxed{529.75}$$

For $M_{12}$, use heights at the midpoint of each subinterval:

$$M_{12}=0.5(11.1875+18.6875+27.6875+38.1875+50.1875+63.6875+78.6875+95.1875+113.1875+132.6875+153.6875+176.1875)=\boxed{479.625}$$

Common Misconceptions

• A Riemann sum is exact. It is an approximation. It only approaches the true definite integral as the number of subintervals grows and the widths shrink toward 0.
• Left always underestimates and right always overestimates. That is only true for increasing functions. For decreasing functions the roles flip.
• The over/underestimate rule for trapezoidal and midpoint sums depends on increasing/decreasing. It depends on concavity, not direction. Trapezoidal overestimates when concave up; midpoint underestimates when concave up.
• Midpoint means averaging the two endpoint heights. Averaging endpoints gives the trapezoidal value. The midpoint sum uses the function value at the middle x of each subinterval.
• All partitions must be uniform. Subintervals can have different widths. With a nonuniform partition, multiply each height by its own width before adding.
• You always need a formula. When a problem gives a table, you can approximate directly from the listed values without ever knowing the underlying function.

Related AP Calculus Guides

• Unit 6 Overview: Integration and Accumulation of Change
• 6.11 Integrating Using Integration by Parts
• 6.1 Integration and Accumulation of Change
• 6.12 Integrating Using Linear Partial Fractions
• 6.4 The Fundamental Theorem of Calculus and Accumulation Functions
• 6.5 Interpreting the Behavior of Accumulation Functions Involving Area