Home » Vector Functions » Arc length and curvature

Collapse menu
Introduction Search the book
1 Analytic Geometry
2 Instantaneous Rate of
Change:
13.3 Arc length and
The Derivative
3 Rules for Finding
curvature
Derivatives [Jump to exercises]
4 Transcendental Functions
5 Curve Sketching Sometimes it is useful to compute the length of a curve in space; for
6 Applications of the example, if the curve represents the path of a moving object, the
Derivative length of the curve between two points may be the distance traveled
7 Integration by the object between two times.
8 Techniques of Integration
9 Applications of Integration
Recall that if the curve is given by the vector function 𝐫 then the
10 Polar Coordinates,
vector Δ𝐫 = 𝐫(𝑡 + Δ𝑡) − 𝐫(𝑡) points from one position on the curve to
Parametric Equations another, as depicted in figure 13.2.1. If the points are close together,
11 Sequences and Series the length of Δ𝐫 is close to the length of the curve between the two
12 Three Dimensions points. If we add up the lengths of many such tiny vectors, placed
13 Vector Functions head to tail along a segment of the curve, we get an approximation to
1. Space Curves the length of the curve over that segment. In the limit, as usual, this
2. Calculus with vector functions sum turns into an integral that computes precisely the length of the
3. Arc length and curvature curve. First, note that
4. Motion along a curve
14 Partial Differentiation |Δ𝐫|
|Δ𝐫| = Δ𝑡 ≈ |𝐫′(𝑡)| Δ𝑡,
Δ𝑡
15 Multiple Integration
16 Vector Calculus when Δ𝑡 is small. Then the length of the curve between 𝐫(𝑎) and 𝐫(𝑏)
is

𝑛−1 𝑛−1 𝑛−1 𝑏

|Δ𝐫|
∫𝑎
|𝐫′(𝑡)| Δ𝑡 = |𝐫′(𝑡)| 𝑑𝑡.
𝑛→∞ ∑ 𝑛→∞ ∑ Δ𝑡 𝑛→∞ ∑
lim |Δ𝐫| = lim Δ𝑡 = lim
𝑖=0 𝑖=0 𝑖=0

(Well, sometimes. This works if between 𝑎 and 𝑏 the segment of

curve is traced out exactly once.)

Example 13.3.1 Let's find the length of one turn of the helix
:
 𝐫 = ⟨cos 𝑡, sin 𝑡, 𝑡⟩ (see figure 13.1.1). We compute
𝐫′ = ⟨− sin 𝑡, cos 𝑡, 1⟩ and |𝐫′| = √‾sin
‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾
2
𝑡 + cos2 𝑡 + ‾1 = √2‾, so the
length is
2𝜋

∫0 √2‾ 𝑑𝑡 = 2√2‾𝜋.

Example 13.3.2 Suppose 𝑦 = ln 𝑥 ; what is the length of this curve

between 𝑥 = 1 and 𝑥 = √3‾?

Although this problem does not appear to involve vectors or three

dimensions, we can interpret it in those terms: let 𝐫(𝑡) = ⟨𝑡, ln 𝑡, 0⟩.
This vector function traces out precisely 𝑦 = ln 𝑥 in the 𝑥 -𝑦 plane.
Then 𝐫′(𝑡) = ⟨1, 1/𝑡, 0⟩ and |𝐫′(𝑡)| = √‾1‾‾‾‾‾‾
+ 1/𝑡‾2 and the desired length
is

√3 ‾‾‾‾‾‾1‾ 1
∫1 √
1+ 𝑑𝑡 = 2 − √2‾ + ln(√2‾ + 1) − ln 3.
𝑡 2 2

(This integral is a bit tricky, but requires only methods we have

learned.) ◻

Notice that there is nothing special about 𝑦 = ln 𝑥 , except that the

resulting integral can be computed. In general, given any 𝑦 = 𝑓(𝑥) ,
we can think of this as the vector function 𝐫(𝑡) = ⟨𝑡, 𝑓(𝑡), 0⟩ . Then
𝐫′(𝑡) = ⟨1, 𝑓 ′ (𝑡), 0⟩ and |𝐫′(𝑡)| = √‾1‾‾‾‾‾‾‾
+ (𝑓 ′ )‾2 . The length of the curve
𝑦 = 𝑓(𝑥) between 𝑎 and 𝑏 is thus

∫𝑎 √
‾1‾‾‾‾‾‾‾‾‾‾
+ (𝑓 ′ (𝑥))‾2 𝑑𝑥.

Unfortunately, such integrals are often impossible to do exactly and

must be approximated.

One useful application of arc length is the arc length

parameterization. A vector function 𝐫(𝑡) gives the position of a
point in terms of the parameter 𝑡 , which is often time, but need not
:
 be. Suppose 𝑠 is the distance along the curve from some fixed
starting point; if we use 𝑠 for the variable, we get 𝐫(𝑠) , the position in
space in terms of distance along the curve. We might still imagine
that the curve represents the position of a moving object; now we get
the position of the object as a function of how far the object has
traveled.

Example 13.3.3 Suppose 𝐫(𝑡) = ⟨cos 𝑡, sin 𝑡, 0⟩. We know that this
curve is a circle of radius 1. While 𝑡 might represent time, it can also
in this case represent the usual angle between the positive 𝑥 -axis and
𝐫(𝑡) . The distance along the circle from (1, 0, 0) to (cos 𝑡, sin 𝑡, 0) is
also 𝑡 —this is the definition of radian measure. Thus, in this case
𝑠 = 𝑡 and 𝐫(𝑠) = ⟨cos 𝑠, sin 𝑠, 0⟩ . ◻

Example 13.3.4 Suppose 𝐫(𝑡) = ⟨cos 𝑡, sin 𝑡, 𝑡⟩. We know that this
curve is a helix. The distance along the helix from (1, 0, 0) to
(cos 𝑡, sin 𝑡, 𝑡) is

𝑡 𝑡 𝑡

∫0 ∫0 ∫0 √
𝑠= |𝐫′(𝑢)| 𝑑𝑢 = √‾cos
‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾
2
𝑢 + sin2 𝑢 + ‾1 𝑑𝑢 = 2‾ 𝑑𝑢 = √2‾𝑡.

Thus, the value of 𝑡 that gets us distance 𝑠 along the helix is 𝑡 = 𝑠/√2‾,
and so the same curve is given by 𝐫̂(𝑠) = ⟨cos(𝑠/√2‾), sin(𝑠/√2‾), 𝑠/√2‾⟩ .
◻

In general, if we have a vector function 𝐫(𝑡) , to convert it to a vector

function in terms of arc length we compute
𝑡

∫𝑎
𝑠= |𝐫′(𝑢)| 𝑑𝑢 = 𝑓(𝑡),

solve 𝑠 = 𝑓(𝑡) for 𝑡 , getting 𝑡 = 𝑔(𝑠) , and substitute this back into 𝐫(𝑡)
to get 𝐫̂(𝑠) = 𝐫(𝑔(𝑠)).

Suppose that 𝑡 is time. By the Fundamental Theorem of Calculus, if

we start with arc length
𝑡

∫𝑎
𝑠(𝑡) = |𝐫′(𝑢)| 𝑑𝑢
:
 and take the derivative, we get

𝑠′(𝑡) = |𝐫′(𝑡)|.

Here 𝑠′(𝑡) is the rate at which the arc length is changing, and we have
seen that |𝐫′(𝑡)| is the speed of a moving object; these are of course
the same.

Suppose that 𝐫(𝑠) is given in terms of arc length; what is |𝐫′(𝑠)| ? It is

the rate at which arc length is changing relative to arc length; it must
be 1! In the case of the helix, for example, the arc length
parameterization is ⟨cos(𝑠/√2‾), sin(𝑠/√2‾), 𝑠/√2‾⟩, the derivative is
⟨− sin(𝑠/√2‾)/√2‾, cos(𝑠/√2‾)/√2‾, 1/√2‾⟩ , and the length of this is

‾sin
‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾ 1‾
√
2 ‾1‾‾‾‾‾1‾
(𝑠/√2‾) cos2 (𝑠/√2‾)
2 √2
+ + = + = 1.
2 2 2

So in general, 𝐫′ is a unit tangent vector.

Given a curve 𝐫(𝑡) , we would like to be able to measure, at various

points, how sharply curved it is. Clearly this is related to how "fast'' a
tangent vector is changing direction, so a first guess might be that we
can measure curvature with |𝐫″(𝑡)| . A little thought shows that this is
flawed; if we think of 𝑡 as time, for example, we could be tracing out
the curve more or less quickly as time passes. The second derivative
|𝐫″(𝑡)| incorporates this notion of time, so it depends not simply on
the geometric properties of the curve but on how quickly we move
along the curve.

Example 13.3.5 Consider 𝐫(𝑡) = ⟨cos 𝑡, sin 𝑡, 0⟩ and

𝐬(𝑡) = ⟨cos 2𝑡, sin 2𝑡, 0⟩. Both of these vector functions represent the
unit circle in the 𝑥 -𝑦 plane, but if 𝑡 is interpreted as time, the second
describes an object moving twice as fast as the first. Computing the
second derivatives, we find |𝐫″(𝑡)| = 1 , |𝐬″(𝑡)| = 4 . ◻

To remove the dependence on time, we use the arc length

parameterization. If a curve is given by 𝐫(𝑠) , then the first derivative
:
 𝐫′(𝑠) is a unit vector, that is, 𝐫′(𝑠) = 𝐓(𝑠) . We now compute the
second derivative 𝐫″(𝑠) = 𝐓′ (𝑠) and use |𝐓′ (𝑠)| as the "official''
measure of curvature, usually denoted 𝜅 .

Example 13.3.6 We have seen that the arc length parameterization

of a particular helix is 𝐫(𝑠) = ⟨cos(𝑠/√2‾), sin(𝑠/√2‾), 𝑠/√2‾⟩ . Computing
the second derivative gives 𝐫″(𝑠) = ⟨− cos(𝑠/√2‾)/2, − sin(𝑠/√2‾)/2, 0⟩
with length 1/2 . ◻

What if we are given a curve as a vector function 𝐫(𝑡) , where 𝑡 is not

arc length? We have seen that arc length can be difficult to compute;
fortunately, we do not need to convert to the arc length
parameterization to compute curvature. Instead, let us imagine that
we have done this, so we have found 𝑡 = 𝑔(𝑠) and then formed
′
𝐫̂(𝑠) = 𝐫(𝑔(𝑠)). The first derivative 𝐫̂ (𝑠) is a unit tangent vector, so it
is the same as the unit tangent vector 𝐓(𝑡) = 𝐓(𝑔(𝑠)). Taking the
derivative of this we get

𝑑 𝑑𝑡
𝐓(𝑔(𝑠)) = 𝐓′ (𝑔(𝑠))𝑔′ (𝑠) = 𝐓′ (𝑡) .
𝑑𝑠 𝑑𝑠

The curvature is the length of this vector:

𝑑𝑡 |𝐓′ (𝑡)| |𝐓′ (𝑡)|

𝜅 = |𝐓′ (𝑡)|| |= = ′ .
𝑑𝑠 |𝑑𝑠/𝑑𝑡| |𝐫 (𝑡)|

(Recall that we have seen that 𝑑𝑠/𝑑𝑡 = |𝐫′(𝑡)| .) Thus we can compute
the curvature by computing only derivatives with respect to 𝑡 ; we do
not need to do the conversion to arc length.

Example 13.3.7 Returning to the helix, suppose we start with the

parameterization 𝐫(𝑡) = ⟨cos 𝑡, sin 𝑡, 𝑡⟩. Then 𝐫′(𝑡) = ⟨− sin 𝑡, cos 𝑡, 1⟩ ,
|𝐫′(𝑡)| = √2‾, and 𝐓(𝑡) = ⟨− sin 𝑡, cos 𝑡, 1⟩/√2‾ . Then
𝐓′ (𝑡) = ⟨− cos 𝑡, − sin 𝑡, 0⟩/√2‾ and |𝐓′ (𝑡)| = 1/√2‾ . Finally,
𝜅 = 1/√2‾/√2‾ = 1/2 , as before. ◻

Example 13.3.8 Consider this circle of radius 𝑎 :

𝐫(𝑡) = ⟨𝑎 cos 𝑡, 𝑎 sin 𝑡, 1⟩ . Then 𝐫′(𝑡) = ⟨−𝑎 sin 𝑡, 𝑎 cos 𝑡, 0⟩ , |𝐫′(𝑡)| = 𝑎 ,
and 𝐓(𝑡) = ⟨−𝑎 sin 𝑡, 𝑎 cos 𝑡, 0⟩/𝑎 . Now 𝐓′ (𝑡) = ⟨−𝑎 cos 𝑡, −𝑎 sin 𝑡, 0⟩/𝑎
:
 and |𝐓′ (𝑡)| = 1 . Finally, 𝜅 = 1/𝑎 : the curvature of a circle is
everywhere the reciprocal of the radius. It is sometimes useful to
think of curvature as describing what circle a curve most resembles
at a point. The curvature of the helix in the previous example is 1/2 ;
this means that a small piece of the helix looks very much like a
circle of radius 2 , as shown in figure 13.3.1. ◻

Figure 13.3.1. A circle with the same curvature as the helix.

Example 13.3.9 Consider 𝐫(𝑡) = ⟨cos 𝑡, sin 𝑡, cos 2𝑡⟩, as shown in

figure 13.2.4. 𝐫′(𝑡) = ⟨− sin 𝑡, cos 𝑡, −2 sin(2𝑡)⟩ and
|𝐫′(𝑡)| = √‾1‾‾‾‾‾‾‾‾‾‾‾ ‾, so
+ 4 sin2 (2𝑡)

⟨ ‾⟩
− sin 𝑡 cos 𝑡 −2 sin 2𝑡
𝐓(𝑡) = , , .
‾1‾‾‾‾‾‾‾‾‾‾‾ ‾ ‾1‾‾‾‾‾‾‾‾‾‾‾ ‾ ‾1‾‾‾‾‾‾‾‾‾‾‾
√ + 4 sin2 (2𝑡) √ + 4 sin2 (2𝑡) √ + 4 sin2 (2𝑡)

Computing the derivative of this and then the length of the resulting
vector is possible but unpleasant. ◻

Fortunately, there is an alternate formula for the curvature that is

:
 often simpler than the one we have:

|𝐫′(𝑡) × 𝐫″(𝑡)|
𝜅= .
|𝐫′(𝑡)|3

Example 13.3.10 Returning to the previous example, we compute

the second derivative 𝐫″(𝑡) = ⟨− cos 𝑡, − sin 𝑡, −4 cos(2𝑡)⟩ . Then the
cross product 𝐫′(𝑡) × 𝐫″(𝑡) is

⟨−4 cos 𝑡 cos 2𝑡 − 2 sin 𝑡 sin 2𝑡, 2 cos 𝑡 sin 2𝑡 − 4 sin 𝑡 cos 2𝑡, 1⟩.

Computing the length of this vector and dividing by |𝐫′(𝑡)|3 is still a

bit tedious. With the aid of a computer we get

√‾48
‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾
cos4 𝑡 − 48 cos2 𝑡 + 17‾
𝜅= .
(−16 cos4 𝑡 + 16 cos2 𝑡 + 1)3/2

Graphing this we get

𝜋 3𝜋
2
𝜋 2

The highest curvature occurs where the curve has its highest and
lowest points, and indeed in the picture these appear to be the most
sharply curved portions of the curve, while the curve is almost a
straight line midway between those points. ◻

Let's see why this alternate formula is correct. Starting with the
definition of 𝐓 , 𝐫′ = |𝐫′|𝐓 so by the product rule 𝐫″ = |𝐫′|′𝐓 + |𝐫′|𝐓′ .
Then by Theorem 12.4.2 the cross product is
′ ″
× = | ′|𝐓 × (| ′ ′
𝐓 + | ′| ′
)
:
 𝐫′ × 𝐫″ = |𝐫′|𝐓 × (|𝐫′|′𝐓 + |𝐫′|𝐓′ )
= |𝐫′|𝐓 × |𝐫′|′𝐓 + |𝐫′|𝐓 × |𝐫′|𝐓′
= |𝐫′||𝐫′|′(𝐓 × 𝐓) + |𝐫′|2 (𝐓 × 𝐓′ )
= |𝐫′|2 (𝐓 × 𝐓′ )

because 𝐓 × 𝐓 = 0 , since 𝐓 is parallel to itself. Then

|𝐫′ × 𝐫″| = |𝐫′|2 |𝐓 × 𝐓′ |

= |𝐫′|2 |𝐓||𝐓′ | sin 𝜃
= |𝐫′|2 |𝐓′ |

using exercise 8 in section 13.2 to see that 𝜃 = 𝜋/2 . Dividing both

sides by |𝐫′|3 then gives the desired formula.

We used the fact here that 𝐓′ is perpendicular to 𝐓 ; the vector

𝐍 = 𝐓′ /|𝐓′ | is thus a unit vector perpendicular to 𝐓 , called the unit
normal to the curve. Occasionally of use is the unit binormal
𝐁 = 𝐓 × 𝐍, a unit vector perpendicular to both 𝐓 and 𝐍.

Exercises 13.3

Sage can help with the somewhat tedious calculation of curvature.

1 t=var('t')
2 r=vector([cos(t),sin(t),cos(2*t)])
3 rp=diff(r,t)
4 rpp=diff(rp,t)
5 k=(norm(rp.cross_product(rpp))/norm(rp)^3).simplify_full
6 show(k)

Evaluate

Ex 13.3.1 Find the length of ⟨3 cos 𝑡, 2𝑡, 3 sin 𝑡⟩, 𝑡 ∈ [0, 2𝜋] . (answer)

Ex 13.3.2 Find the length of ⟨𝑡2 , 2, 𝑡3 ⟩ , 𝑡 ∈ [0, 1]. (answer)

Ex 13.3.3 Find the length of ⟨𝑡2 , sin 𝑡, cos 𝑡⟩ , 𝑡 ∈ [0, 1]. (answer)

Ex 13.3.4 Find the length of the curve 𝑦 = 𝑥3/2 , 𝑥 ∈ [1, 9] . (answer)

:
 Ex 13.3.5 Set up an integral to compute the length of ⟨cos 𝑡, sin 𝑡, 𝑒𝑡 ⟩ ,
𝑡 ∈ [0, 5]. (It is tedious but not too difficult to compute this integral.)
(answer)

Ex 13.3.6 Find the curvature of ⟨𝑡, 𝑡2 , 𝑡⟩ . (answer)

Ex 13.3.7 Find the curvature of ⟨𝑡, 𝑡2 , 𝑡2 ⟩ . (answer)

Ex 13.3.8 Find the curvature of ⟨𝑡, 𝑡2 , 𝑡3 ⟩ . (answer)

Ex 13.3.9 Find the curvature of 𝑦 = 𝑥4 at (1, 1) . (answer)

: