Why Zero Net Change May Hide Large Changes

Equal Endpoints Do Not Mean Nothing Happened

A quantity can rise, fall, and return to its starting value. Its final value then matches its initial value, but it has still changed substantially along the way. A definite integral of a rate records the net change over an interval: increases count positively and decreases count negatively. Opposing contributions can therefore cancel.

This distinction matters whenever an integral represents a changing quantity, such as position changing at a velocity, water volume changing at an inflow rate, or a balance changing at a deposit or withdrawal rate. A net change of zero describes the difference between the endpoints. It does not, by itself, describe what happened between them.

Signed Accumulation and Cancellation

For a rate function \(r\), the definite integral counts the area above the horizontal axis positively and the area below it negatively. If \(r\) is continuous, or is piecewise continuous with only finitely many discontinuities, its signed accumulation can be split at points where its sign changes.

Thus, if the contribution on one part is positive and the contribution on another part is negative, the total can be zero even when neither contribution is zero. In area language, the total area above the axis can equal the magnitude of the total area below it.

Suppose \(Q\) is a quantity whose rate is \(r\), so \(Q'(t)=r(t)\), and suppose \(r\) is continuous on \([a,b]\). The Fundamental Theorem of Calculus gives:

If the integral is zero, then \(Q(b)=Q(a)\). This equation compares only the endpoint values. It does not say that \(Q(t)\) stays equal to that common value at every intermediate time.

Why Cancellation Works

The reason is the additivity of definite integrals. When an interval is split at a point \(c\), the total signed accumulation is the sum of the accumulations on the two smaller intervals. A positive contribution and an equal-magnitude negative contribution have sum zero.

For example, if the rate is positive on \([a,c]\) and negative on \([c,b]\), then zero net accumulation means:

The left side is positive and the right side is also positive when \(r(t)<0\) on \([c,b]\). They represent equal amounts of change in opposite directions. Their sum as signed contributions is zero.

Worked Example: A Rate Reverses Direction

Zero Net Change Versus Total Movement

When a rate can be positive and negative, it is useful to distinguish two questions:

• What is the net change? Use the signed integral \(\int_a^b r(t)\,dt\).
• How much change accumulated in either direction? For a piecewise continuous rate, use \(\int_a^b |r(t)|\,dt\).

The absolute value makes contributions in either direction count positively. This integral measures the total amount of movement described by the rate, rather than the endpoint difference. In the previous example, the net change was \(0\) units, while the total movement was \(8+8=16\) units. The units remain units because integrating a rate in units per minute over time in minutes gives units.

The total movement is not generally the same as the difference between the largest and smallest values of the quantity. A quantity could reverse direction several times, accumulating movement on each segment. The signed integral records the net result; the absolute-value integral records the sum of the magnitudes of those changes.

Worked Example: Equal Triangles Cancel

A Complete Reason for the Endpoint Result

Here is the precise link between the rate and the quantity. If \(Q\) is differentiable on an interval and \(Q'(t)=r(t)\), with \(r\) continuous on \([a,b]\), the Fundamental Theorem of Calculus gives:

Therefore, if \(\int_a^b r(t)\,dt=0\), then \(Q(b)-Q(a)=0\), so \(Q(b)=Q(a)\). The conclusion concerns the endpoints. To see why it does not force \(Q\) to be constant inside the interval, consider \(r(t)=t-2\) from the example: the rate is nonzero except at \(t=2\), and the accumulated change reaches \(-2\) units before returning to zero.

There is an important contrast. If \(r\) is continuous and nonnegative throughout \([a,b]\), then a zero integral does force \(r(t)=0\) at every point. To prove this, suppose instead that \(r(c)>0\) for some \(c\). By continuity, \(r\) stays positive and bounded away from zero on a small interval around \(c\) within \([a,b]\). Its area on that interval is then positive. Since \(r\ge0\) everywhere else, the integral over the full interval must be positive, contradicting \(\int_a^b r(t)\,dt=0\). The same reasoning applies if \(r\le0\), by considering \(-r\). Thus, cancellation requires contributions of opposing signs; a continuous rate that keeps one sign cannot have a zero integral unless it is zero throughout.

Worked Example: A Reservoir Returns to Its Starting Volume

A Common Mistake

A frequent error is to interpret a zero integral as saying that the rate is zero throughout the interval.

This reasoning confuses a signed total with the values being added. In fact, \(t-2<0\) for \(0\le t<2\) and \(t-2>0\) for \(2<t\le4\). The negative triangular contribution is \(-2\) units, and the positive triangular contribution is \(2\) units. Their sum is zero, although the rate is nonzero at nearly every time.

The corrected interpretation is: the total increase and total decrease balance, so the endpoint values are equal. To determine whether the quantity changed in between, examine the rate’s signs and the accumulation on subintervals.