Approximating areas with Riemann sums — AP Calculus BC Study Guide

For: AP Calculus BC candidates sitting AP Calculus BC.

Covers: Left Riemann sums, right Riemann sums, midpoint Riemann sums, trapezoidal sums, sigma notation for Riemann sums, over/under approximation for monotonic functions, equal and unequal subinterval area approximation on bounded intervals.

You should already know: 1. Evaluating finite sums and sigma notation. 2. Properties of monotonic (increasing/decreasing) functions and concavity. 3. Basic area formulas for rectangles and trapezoids.

A note on the practice questions: All worked questions in the "Practice Questions" section below are original problems written by us in the AP Calculus BC style for educational use. They are not reproductions of past College Board / Cambridge / IB papers and may differ in wording, numerical values, or context. Use them to practise the technique and cross-check with the mark schemes for grading.

1. What Is Approximating areas with Riemann sums?

Approximating areas with Riemann sums is the foundational numerical method for finding the area bounded by a curve (y=f(x)), the (x)-axis, and two vertical lines (x=a) and (x=b), developed before we define the exact definite integral. For the AP Calculus BC CED, this topic makes up roughly 10-12% of Unit 6 (Integration and Accumulation of Change) exam weight, and it appears in both multiple-choice (MCQ) and free-response (FRQ) sections.

The core intuition is simple: when you cannot find an exact antiderivative to calculate area, you split the interval ([a,b]) into (n) smaller subintervals, approximate the area over each subinterval with a simple geometric shape (rectangle for standard Riemann sums, trapezoid for trapezoidal sums), then add all these small areas together. As the number of subintervals (n) increases, the width of each subinterval shrinks, and the approximation gets arbitrarily close to the true area. This process is sometimes called the rectangle approximation method (RAM), and it directly motivates the limit definition of the definite integral. On the AP exam, you will be asked to compute approximations, write them in sigma notation, classify them as over- or under-estimates, or apply them to real-world contexts.

2. Riemann Sums with Equal Subintervals

The most common case you will encounter on the AP exam is Riemann sums with equal-width subintervals. For an interval ([a,b]) split into (n) subintervals, the width of every subinterval is given by: $$\Delta x=\frac{b-a}{n}$$ We label the endpoints of the subintervals (x_i = a + i\Delta x) for (i=0,1,2,...,n), so the (i)-th subinterval is ([x_{i-1}, x_i]). There are four standard types of equal-subinterval Riemann sums, each using a different point to find the height of the approximating shape:

1. Left Riemann Sum: Uses the left endpoint of each subinterval for height: (L_n = \Delta x \sum_{i=1}^n f(x_{i-1}))
2. Right Riemann Sum: Uses the right endpoint of each subinterval for height: (R_n = \Delta x \sum_{i=1}^n f(x_i))
3. Midpoint Riemann Sum: Uses the midpoint of each subinterval for height: (M_n = \Delta x \sum_{i=1}^n f\left(\frac{x_{i-1}+x_i}{2}\right))
4. Trapezoidal Sum: Uses a trapezoid to approximate area over each subinterval, which simplifies to: (T_n = \frac{\Delta x}{2}\left[f(x_0) + 2f(x_1) + ... + 2f(x_{n-1}) + f(x_n)\right])

For monotonic functions, we can immediately classify a sum as over or under approximate: if (f(x)) is increasing, left sums are underestimates and right sums are overestimates; if (f(x)) is decreasing, this pattern reverses. For midpoint and trapezoidal sums, concavity determines bias: concave up functions have trapezoidal sums as overestimates and midpoint sums as underestimates, with the opposite for concave down.

Worked Example

Approximate the area under (f(x) = x^2 + 1) on the interval ([0, 2]) using 4 equal subintervals. Calculate (a) Left Riemann Sum, (b) Right Riemann Sum, (c) Midpoint Riemann Sum.

1. First calculate (\Delta x): (\Delta x = \frac{2-0}{4} = 0.5).
2. List all endpoints: (x_0=0, x_1=0.5, x_2=1.0, x_3=1.5, x_4=2.0), and calculate (f(x)) at each: (f(0)=1, f(0.5)=1.25, f(1)=2, f(1.5)=3.25, f(2)=5).
3. (a) Left Sum uses left endpoints (x_0) to (x_3): $$L_4=0.5\left(1+1.25+2+3.25\right)=0.5(7.5)=3.75$$
4. (b) Right Sum uses right endpoints (x_1) to (x_4): $$R_4=0.5\left(1.25+2+3.25+5\right)=0.5(11.5)=5.75$$
5. (c) Midpoints of the subintervals are (0.25, 0.75, 1.25, 1.75). Calculate (f) at each: (f(0.25)=1.0625, f(0.75)=1.5625, f(1.25)=2.5625, f(1.75)=4.0625). Multiply by (\Delta x): $$M_4=0.5\left(1.0625+1.5625+2.5625+4.0625\right)=0.5(9.25)=4.625$$

Exam tip: On the AP exam, always explicitly label all endpoints (x_0) through (x_n) before calculating. This eliminates the common mistake of mixing up left and right endpoints, which is one of the most frequent sources of lost points on Riemann sum problems.

3. Riemann Sums in Sigma Notation

AP exams regularly ask you to write a Riemann sum in sigma notation, or identify what type of sum is given in sigma form. The general form of any equal-subinterval Riemann sum is: $$\text{Area}\approx \Delta x\sum _{i=1}^{n}f(x_i^{\ast })$$ where (x_i^) is the sample point (left, right, midpoint) in the (i)-th subinterval. To write the sum, you only need to express (x_i^) in terms of (i), substitute into (f), and confirm the index matches the starting/ending points. A common trick on MCQ is shifting the index of the sum to test if you can still recognize the type of Riemann sum; always check the first and last term of the sum to confirm which endpoints you are using.

Worked Example

Write the right Riemann sum for (f(x) = \ln x) on the interval ([1, 5]) with 8 equal subintervals in sigma notation (do not evaluate the sum). Then identify what type of sum is given by (\frac{1}{2} \sum_{i=0}^7 \ln\left(1 + \frac{i}{2}\right)).

1. Calculate (\Delta x): (\Delta x = \frac{5-1}{8} = \frac{1}{2}).
2. For a right Riemann sum with (n=8), the right endpoint of the (i)-th subinterval (for (i=1) to (8)) is (x_i = a + i\Delta x = 1 + i\left(\frac{1}{2}\right)).
3. Substitute into the general formula to get the right sum: $$\frac{1}{2}\sum _{i=1}^8\ln \left(1+\frac{i}{2}\right)$$
4. For the second sum, the index runs from (i=0) to (i=7). The first sample point is (i=0): (1 + 0 = 1), which is the left endpoint of the first subinterval ([1, 1.5]). The last sample point is (i=7): (1 + \frac{7}{2} = 4.5), which is the left endpoint of the last subinterval ([4.5, 5]). This is a left Riemann sum with 8 equal subintervals.

Exam tip: Always check the starting index of the sigma: a sum starting at (i=0) and ending at (n-1) for (n) subintervals is always a left sum, while a sum starting at (i=1) and ending at (n) is always a right sum.

4. Riemann Sums for Unequal Subintervals

Many AP FRQ problems give you a table of function values at non-equally spaced points, and ask you to calculate a Riemann sum from the table. For unequal subintervals, each subinterval between consecutive points (x_{i-1}) and (x_i) has its own width: (\Delta x_i = x_i - x_{i-1}). You calculate the area for each subinterval separately, multiply the height by the subinterval width, then add all areas. This is extremely common in real-world context problems, where measurements are taken at irregular time intervals.

Worked Example

The table below gives values of a decreasing function (v(t)), the velocity of a car in miles per hour, at different times (t) in hours:

Approximate the total distance traveled by the car from (t=0) to (t=6) using a left Riemann sum, and state if the approximation is an over or under estimate.

1. First list the subintervals and calculate each width separately: the subintervals are ([0,1], [1,3], [3,6]), so (\Delta x_1 = 1-0=1), (\Delta x_2=3-1=2), (\Delta x_3=6-3=3).
2. A left Riemann sum uses the left endpoint of each subinterval for velocity, so we use (v(0)=60), (v(1)=52), (v(3)=45).
3. Calculate the total approximate distance: $$\text{Distance}\approx (60\cdot 1)+(52\cdot 2)+(45\cdot 3)=60+104+135=299$$
4. Since (v(t)) is decreasing, the left endpoint of each interval is the largest value of (v(t)) on that interval, so this approximation is an overestimate of the true distance.

Exam tip: For table problems, never assume all widths are equal. Always list each width first before starting your calculation to avoid losing points for assuming uniform spacing.

Common Pitfalls (and how to avoid them)

• Wrong move: Using (n) (the number of subintervals) instead of (\frac{b-a}{n}) for (\Delta x), e.g., using (\Delta x=4) for (n=4) on ([0,2]). Why: Confusing the count of subintervals with their width, especially when writing sigma notation. Correct move: Always calculate and label (\Delta x) explicitly before starting any calculation.
• Wrong move: Reversing the over/under classification for left/right sums, e.g., claiming a left sum for an increasing function is an overestimate. Why: Blind memorization without intuition leads to reversal. Correct move: Draw a 10-second rough sketch of the function and rectangles to see if the rectangles lie above or below the curve.
• Wrong move: For the trapezoidal rule, forgetting the (\frac{\Delta x}{2}) leading factor or forgetting to double the middle terms. Why: Misremembering the formula. Correct move: Derive the trapezoidal sum as the average of the left and right sum: (T_n = \frac{L_n + R_n}{2}), which automatically gives the correct coefficients.
• Wrong move: For unequal subinterval table problems, using the same average width for all subintervals even when spacing is uneven. Why: Habit from equal-subinterval problems leads to automatic assumption. Correct move: Always list all subinterval widths first before calculating the sum.
• Wrong move: Starting a right Riemann sum at (i=0) instead of (i=1), leading to including the left endpoint of the entire interval and missing the right endpoint. Why: Confusion about index numbering. Correct move: Write out the first and last term of the sum to confirm the endpoints match the interval requirements.

Practice Questions (AP Calculus BC Style)

Question 1 (Multiple Choice)

Let (f(x)) be a concave down increasing function on the interval ([2, 6]). A midpoint Riemann sum with 4 equal subintervals is used to approximate the area under (f(x)). Which of the following statements is true? A) The midpoint sum is an underestimate of the true area B) The midpoint sum is an overestimate of the true area C) The midpoint sum is equal to the true area D) The relationship between the midpoint sum and the true area cannot be determined from the given information

Worked Solution: Bias in midpoint and trapezoidal sums is determined by the concavity of the function, not whether it is increasing or decreasing. For a concave down function, the curve lies below the tangent line at the midpoint of any subinterval, so the midpoint rectangle (with height equal to the function value at the midpoint) will have a larger area than the true area under the curve over the subinterval. This means the full midpoint sum will be an overestimate of the total true area. The correct answer is B.

Question 2 (Free Response)

Consider the function (g(x) = \sqrt{x + 2}) on the interval ([0, 4]). (a) Write the midpoint Riemann sum for (g(x)) on ([0, 4]) with 4 equal subintervals in sigma notation. (b) Calculate the value of this midpoint Riemann sum, round your answer to two decimal places. (c) State whether your approximation is an overestimate or underestimate of the true area under (g(x)) on ([0, 4]). Justify your answer.

Worked Solution: (a) First calculate (\Delta x = \frac{4-0}{4} = 1). The (i)-th subinterval for (i=1) to 4 is ([i-1, i]), so its midpoint is (i - 0.5). Substituting into the midpoint sum formula gives: $$\sum _{i=1}^4\sqrt{(i-0.5)+2}=\sum _{i=1}^4\sqrt{i+1.5}$$

(b) Evaluate each term of the sum: (i=1: \sqrt{2.5} \approx 1.5811), (i=2: \sqrt{3.5} \approx 1.8708), (i=3: \sqrt{4.5} \approx 2.1213), (i=4: \sqrt{5.5} \approx 2.3452) Adding these gives (1.5811 + 1.8708 + 2.1213 + 2.3452 = 7.9184 \approx 7.92).

(c) Calculate the second derivative to find concavity: (g'(x) = \frac{1}{2}(x+2)^{-1/2}), (g''(x) = -\frac{1}{4}(x+2)^{-3/2}), which is negative for all (x) in ([0,4]). This means (g(x)) is concave down on the interval, so the midpoint Riemann sum is an overestimate of the true area.

Question 3 (Application / Real-World Style)

A population biologist measures the growth rate of a deer population (r(t)), in deer per year, where (t) is time in years since the start of the study. The table below gives measurements of (r(t)) at different times:

Use a trapezoidal sum to approximate the total change in the deer population over the 10-year study. Give your answer as a whole number, and interpret the result in context.

Worked Solution: First, calculate the width of each subinterval: (\Delta x_1 = 2-0=2), (\Delta x_2=5-2=3), (\Delta x_3=8-5=3), (\Delta x_4=10-8=2). For a trapezoidal sum on unequal subintervals, the area of each trapezoid is (\frac{r(t_{i-1}) + r(t_i)}{2} \cdot \Delta x_i). Calculate each term:

• ([0,2]: \frac{12+18}{2} \cdot 2 = 30)
• ([2,5]: \frac{18+25}{2} \cdot 3 = 64.5)
• ([5,8]: \frac{25+21}{2} \cdot 3 = 69)
• ([8,10]: \frac{21+15}{2} \cdot 2 = 36)

Adding all terms gives (30 + 64.5 + 69 + 36 = 199.5 \approx 200). In context, this means the deer population increased by approximately 200 deer over the 10-year study.

Quick Reference Cheatsheet

What's Next

This topic is the foundation for all of integration in AP Calculus BC. Immediately after learning Riemann sums, you will move to the definition of the definite integral as the limit of Riemann sums, which formalizes the idea that the true area is the limit of these approximations as the number of subintervals grows to infinity. Without a solid understanding of how Riemann sums approximate area, the connection between area and the definite integral will remain abstract, making it harder to master the Fundamental Theorem of Calculus and accumulation functions, which are the core of Unit 6.

Riemann sum approximation also directly leads to more advanced numerical integration methods covered in BC, including Simpson's rule, which gives a more accurate approximation for area when an exact antiderivative is not available. All integration applications, from finding area between curves to calculating volumes of revolution, build on the core idea of approximating with small geometric shapes that you learned here.

Follow-on topics:

• The definite integral as the limit of Riemann sums
• The Fundamental Theorem of Calculus
• Numerical integration: Simpson's rule