Understanding Accumulation, Riemann Sums, and the Definite Integral (AP Calculus AB Unit 6)

Exploring Accumulations of Change

What “accumulation” means in calculus

In everyday language, to accumulate means to add up many small contributions over time or over distance. In calculus, accumulation is the same idea, but the “small contributions” come from a rate of change that varies continuously.

For example, if a car’s velocity changes over time, you can’t find the total change in position by doing a single multiplication unless the velocity is constant. Instead, you imagine breaking time into tiny intervals. Over a tiny interval, velocity is almost constant, so distance traveled is approximately:

\text{(velocity)}\times\text{(time interval)}

Then you add these tiny distance pieces over the whole time period. That add-up process is accumulation.

The key mindset is:

• A rate (like velocity, water flow, or marginal cost) tells you “how fast” something changes.
• An accumulated change (like displacement, total volume, or total cost) tells you “how much changed overall.”

Calculus provides a precise way to turn rates into total change.

Rate vs. total change (and why units matter)

A reliable way to understand accumulation is through units.

Suppose r(t) is a rate measured in “units per second.” If you multiply by a small time width \Delta t (in seconds), you get:

r(t)\Delta t

The seconds cancel, leaving “units.” That means each product r(t)\Delta t represents a small amount of the accumulated quantity.

This unit-checking is not just a detail; it prevents common mistakes such as confusing a rate with a total.

Accumulation from a graph: area as “total change”

One of the most important ideas in this section is that the total accumulated change is represented by area under a rate curve, but you must interpret “area” correctly.

If r(x) is a rate as a function of x, then the accumulation from x=a to x=b is the **signed area** between the graph of r and the x-axis:

• Where r(x) > 0, the accumulation increases.
• Where r(x) < 0, the accumulation decreases.

So in calculus you often distinguish:

• Net change: adds positive and negative contributions (signed area).
• Total change: adds magnitudes only (area with absolute value, often requiring piecewise work).

Example 1: Net change from a rate graph (conceptual)

Imagine a rate function r(t) (in gallons per minute) describing water flowing into a tank:

• From t=0 to t=4, the graph is above the axis with area 10.
• From t=4 to t=6, the graph is below the axis with area magnitude 3.

Then the net amount of water added is:

10 - 3 = 7

The interpretation is “we gained 10 gallons, then lost 3 gallons,” for a net gain of 7 gallons.

A common mistake is to say the total added is 10+3 without noticing the rate was negative on the second interval. That would be “total variation,” not net change.

Accumulation functions: building a new function by accumulating

Sometimes you don’t just want a single accumulated value from a to b. You want a function that tells you the accumulation up to any input value. This leads to an accumulation function (also called an accumulation of change function).

A typical definition looks like:

A(x)=\int_a^x f(t)\,dt

Even before you learn all the rules for evaluating integrals, the meaning is clear: A(x) is “the accumulated amount from t=a up to t=x.” The variable inside the integral, t, is a _dummy variable_; it just labels the input for the rate function while x tells you where to stop accumulating.

Two important interpretation skills:

1. Initial value:

A(a)=\int_a^a f(t)\,dt=0

Accumulating over an interval of zero length gives zero.

2. Changing the upper bound changes the accumulation: as x increases, you add more of the rate function’s contributions.

Example 2: Interpreting an accumulation function with units

Suppose v(t) is velocity in meters per second. Define:

s(t)=\int_0^t v(u)\,du

Then s(t) has units:

(\text{meters per second})\times(\text{seconds})=\text{meters}

So s(t) represents **displacement** from time 0 to time t.

If you want distance traveled, you would need to account for direction changes (velocity negative), which often means integrating |v(u)| or splitting into intervals where the sign of v is known.

Exam Focus

• Typical question patterns:
  • You are given a rate function (often via a graph or table) and asked for the net change on an interval by interpreting signed area.
  • You define an accumulation function like A(x)=\int_a^x f(t)\,dt and are asked to interpret A(c) in context (including units).
  • You are asked to distinguish net change vs. total change (often when the rate crosses the axis).
• Common mistakes:
  • Treating negative rate values as positive area when the question asks for net change.
  • Ignoring units, leading to interpreting a rate as a total or vice versa.
  • Confusing the dummy variable and the bound variable (for example, thinking \int_0^x f(x)\,dx is “wrong” just because the same letter appears; it is acceptable notation, but conceptually the inside variable is just a placeholder).

Approximating Areas with Riemann Sums

Why approximation is needed

In many realistic situations, you know a rate function only through measurements (a table) or you have a function whose exact antiderivative is difficult to find. Even when an exact integral exists, the foundational idea of the definite integral is built from approximation: you approximate area by simple shapes, then take a limit as the shapes get thinner.

A Riemann sum is the standard way to do this approximation.

The core geometric idea: rectangles on subintervals

To approximate the accumulated change of f(x) on [a,b], you:

1. Split [a,b] into n subintervals.
2. On each subinterval, pick a representative x-value.
3. Make a rectangle whose width is the subinterval width and whose height is the function value at that representative point.
4. Add the rectangle areas.

If the subintervals have equal width, the width is:

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

Then each rectangle’s area looks like:

f(x_i^*)\Delta x

where x_i^* is the chosen sample point in the ith subinterval.

Left, right, and midpoint sums (and what they tend to do)

The main named Riemann sums differ only in how they choose x_i^*.

• Left Riemann sum: use the left endpoint of each subinterval.
• Right Riemann sum: use the right endpoint of each subinterval.
• Midpoint Riemann sum: use the midpoint of each subinterval.

These choices affect whether you tend to overestimate or underestimate. A useful rule of thumb:

• If f is increasing on [a,b], the left sum tends to underestimate and the right sum tends to overestimate.
• If f is decreasing on [a,b], the left sum tends to overestimate and the right sum tends to underestimate.

Midpoint sums often give better accuracy because the midpoint height can “balance” the variation of the function on each subinterval, but it is still an approximation.

Be careful: these over/under statements are not magic; they rely on monotonic behavior (consistently increasing or decreasing). If f changes direction, you can’t apply the rule globally without analyzing intervals.

Signed area still matters in approximations

When f(x) is below the axis, rectangle areas computed as f(x_i^*)\Delta x will be negative. That is not an error; it matches the idea of net accumulation.

If a question asks for “area between the curve and the axis,” then you are not finding a signed quantity. You would need to use absolute value or split into parts where f keeps a constant sign.

Worked example: approximating an integral with left and right sums

Approximate the net area (net accumulation) under f(x)=x^2 on [0,2] using n=4 subintervals.

1. Compute the width:

\Delta x=\frac{2-0}{4}=\frac{1}{2}

2. List the partition points:

x_0=0,\ x_1=\frac{1}{2},\ x_2=1,\ x_3=\frac{3}{2},\ x_4=2

Left sum uses x_0,x_1,x_2,x_3:

L_4=\left(f(0)+f\left(\frac{1}{2}\right)+f(1)+f\left(\frac{3}{2}\right)\right)\Delta x

Compute function values:

f(0)=0
f\left(\frac{1}{2}\right)=\frac{1}{4}
f(1)=1
f\left(\frac{3}{2}\right)=\frac{9}{4}

Add and multiply:

L_4=\left(0+\frac{1}{4}+1+\frac{9}{4}\right)\left(\frac{1}{2}\right)=\left(\frac{14}{4}\right)\left(\frac{1}{2}\right)=\frac{7}{4}=1.75

Right sum uses x_1,x_2,x_3,x_4:

R_4=\left(f\left(\frac{1}{2}\right)+f(1)+f\left(\frac{3}{2}\right)+f(2)\right)\Delta x

Compute f(2)=4 and add:

R_4=\left(\frac{1}{4}+1+\frac{9}{4}+4\right)\left(\frac{1}{2}\right)=\left(\frac{30}{4}\right)\left(\frac{1}{2}\right)=\frac{15}{4}=3.75

Because f(x)=x^2 is increasing on [0,2], it makes sense that:

L_4 < \int_0^2 x^2\,dx < R_4

The exact value (from antiderivatives) is \frac{8}{3}, which is indeed between 1.75 and 3.75. On an exam, you might not need the exact value; the point is constructing and interpreting the approximations.

Worked example: midpoint sum

Using the same function and interval, the midpoints of the four subintervals are:

\frac{1}{4},\ \frac{3}{4},\ \frac{5}{4},\ \frac{7}{4}

So the midpoint approximation is:

M_4=\left(f\left(\frac{1}{4}\right)+f\left(\frac{3}{4}\right)+f\left(\frac{5}{4}\right)+f\left(\frac{7}{4}\right)\right)\Delta x

Compute:

f\left(\frac{1}{4}\right)=\frac{1}{16}
f\left(\frac{3}{4}\right)=\frac{9}{16}
f\left(\frac{5}{4}\right)=\frac{25}{16}
f\left(\frac{7}{4}\right)=\frac{49}{16}

Add:

M_4=\left(\frac{1+9+25+49}{16}\right)\left(\frac{1}{2}\right)=\left(\frac{84}{16}\right)\left(\frac{1}{2}\right)=\frac{21}{8}=2.625

This is much closer to \frac{8}{3}\approx 2.6667.

Approximating from tables (a very common AP skill)

Often you are given a table of values for f(x) at equally spaced x-values. Then:

• A left sum uses all but the last table value.
• A right sum uses all but the first table value.
• A midpoint sum may require midpoints that are not in the table unless the table already provides them.

The key is to match each rectangle to a subinterval width and the correct sampled height.

Exam Focus

• Typical question patterns:
  • Given a table of values, approximate \int_a^b f(x)\,dx using left, right, or midpoint sums for a stated number of subintervals.
  • Given a graph, approximate the integral by counting/estimating geometric areas or by interpreting rectangles.
  • Compare two approximations (for example, explain whether one is an overestimate given that f is increasing).
• Common mistakes:
  • Using the wrong endpoints (including both endpoints in a left sum, or forgetting which ones a right sum uses).
  • Forgetting the factor \Delta x, especially with table problems.
  • Treating a signed-area question as an “always positive area” question, or vice versa.

Riemann Sums, Summation Notation, and Definite Integral Notation

From “add up rectangles” to compact notation

When you write out a Riemann sum in full, it can be long:

f(x_1^*)\Delta x + f(x_2^*)\Delta x + \cdots + f(x_n^*)\Delta x

Mathematicians use summation notation to write this more compactly.

Building the partition and sample points

Start with an interval [a,b] and split it into n equal pieces.

• Width:

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

• Partition points:

x_i=a+i\Delta x

for integers i=0,1,2,\dots,n.

Then you choose sample points x_i^* inside each subinterval [x_{i-1},x_i]. The choice determines left, right, midpoint, or another method.

• Left endpoint sample for the ith rectangle:

x_i^*=x_{i-1}=a+(i-1)\Delta x

• Right endpoint sample:

x_i^*=x_i=a+i\Delta x

• Midpoint sample:

x_i^*=\frac{x_{i-1}+x_i}{2}=a+\left(i-\frac{1}{2}\right)\Delta x

Summation notation for a general Riemann sum

A general Riemann sum for f(x) on [a,b] is written:

\sum_{i=1}^{n} f(x_i^*)\Delta x

Read it as: “sum from i=1 to n of f(x_i^*)\Delta x.”

What each part means:

• i is the index that counts rectangles.
• n is the number of rectangles.
• x_i^* is the chosen input in the ith subinterval.
• \Delta x is the common width.

A very common AP task is to translate between words, a sigma expression, and an integral.

Definite integral notation: the limit of Riemann sums

The definite integral is defined as the limit of Riemann sums as the rectangles get infinitely thin.

If the limit exists, then:

\int_a^b f(x)\,dx=\lim_{n\to\infty} \sum_{i=1}^{n} f(x_i^*)\Delta x

This equation is the bridge between approximation and exact accumulation:

• The summation represents an approximation with finitely many rectangles.
• The limit represents what happens when the partition becomes infinitely fine.

Even though later you will compute many definite integrals using antiderivatives (Fundamental Theorem of Calculus), it is essential to remember that the integral’s meaning is accumulation (signed area) defined by this limit.

Common equivalent forms you should recognize

Because there are many ways to choose sample points, you may see different but equivalent Riemann sum forms.

Notation reference table

Translating a sigma expression into an integral (and back)

This is a high-value skill: recognizing the interval [a,b], the width \Delta x, and the sample point formula.

Example 1: Identify the integral represented by a limit

Consider:

\lim_{n\to\infty} \sum_{i=1}^{n} \left(3+\frac{2i}{n}\right)^2 \cdot \frac{2}{n}

Step 1: Match \Delta x.

The factor outside is \frac{2}{n}, so:

\Delta x=\frac{2}{n}

That suggests b-a=2.

Step 2: Identify the sample point x_i.

Inside the function you see:

3+\frac{2i}{n}=3+i\cdot\frac{2}{n}=3+i\Delta x

So the sample points match right endpoints for an interval starting at a=3.

Step 3: Determine the interval.

If a=3 and b-a=2, then b=5.

Step 4: Write the integral.

\int_3^5 x^2\,dx

A common misconception here is to think the integrand must literally be \left(3+\frac{2i}{n}\right)^2. That expression is the function evaluated at the sample point; the underlying function is f(x)=x^2.

Example 2: Write a Riemann sum for a given integral using midpoints

Write a midpoint Riemann sum with n equal subintervals for:

\int_1^7 \sqrt{x}\,dx

Step 1: Compute width.

\Delta x=\frac{7-1}{n}=\frac{6}{n}

Step 2: Midpoints are:

x_i^*=1+\left(i-\frac{1}{2}\right)\Delta x

So:

x_i^*=1+\left(i-\frac{1}{2}\right)\frac{6}{n}

Step 3: Build the sum.

\sum_{i=1}^{n} \sqrt{1+\left(i-\frac{1}{2}\right)\frac{6}{n}}\cdot\frac{6}{n}

Two frequent errors are (1) using i instead of i-\frac{1}{2} for midpoints, and (2) forgetting to multiply by \Delta x.

Interpreting definite integrals in context (beyond “area under a curve”)

“Area under a curve” is a helpful mental image, but AP Calculus AB often expects you to interpret integrals as accumulated quantities in context.

If r(t) is a rate of change of some quantity Q with respect to time, then:

\int_{t=a}^{t=b} r(t)\,dt

represents the net change in Q from t=a to t=b.

Examples of common interpretations:

• If v(t) is velocity, then \int v(t)\,dt is displacement.
• If F(t) is flow rate (volume per time), then \int F(t)\,dt is total volume added (net).
• If MC(x) is marginal cost (dollars per item), then integrating it over a production interval gives the change in total cost.

The through-line is always the same: rate times small input-width gives small change; summing gives total change.

Memory aid: “rate times width”

When deciding whether an expression is set up correctly, ask yourself: “Does it look like rate times width?”

• Riemann sum term: f(x_i^*)\Delta x
• Integral: \int_a^b f(x)\,dx

Both represent the same structure: add up lots of rate times width pieces.

Exam Focus

• Typical question patterns:
  • Convert a Riemann sum (sigma notation or a written sum) into a definite integral, identifying a, b, \Delta x, and the sample point type.
  • Create a Riemann sum (often with left/right/midpoint) for a given integral and a specified number of subintervals.
  • Interpret \int_a^b f(x)\,dx in a real-world context with correct units and meaning.
• Common mistakes:
  • Misidentifying \Delta x when it is written as a factor like \frac{b-a}{n} or a decimal.
  • Confusing right endpoint form a+i\Delta x with left endpoint form a+(i-1)\Delta x.
  • Treating the integral as always positive “area,” even when the integrand is negative on part of the interval (net change vs. geometric area confusion).