MA4545: Applied Differential Geometry

Lecture 1: Curves

Course Material: This course covers the basic theory of curves and surfaces in the 3-dimensional Euclidean
space. It provides students with an introduction to the subject of differential geometry.
Assessment Tasks: 28% Midterm Exam ((In class, two hours, the exact date will be announced later));
12% Two Assignments (6% each and will be announced via the Canvas); 60% Examination
Office Hour: 15:00 to 17:00, every Monday

1.1 Suggested Textbooks

• M. do Carmo, 1976, Differential Geometry of Curves and Surfaces. Prentice-Hall.
Remark: the lecture notes follow it and there are many examples and exercises in it.
• M. M. Lipschutz, 1970, Schaum’s Outline of Theory and Problems of Differential Geometry . McGraw-
Hill.

1.2 Contents
1. Curves
Parametrized curves – Regular curves – Vector product – Local theory
2. Regular Surfaces
Definitions – Change of parameters– Tangent plane – First fundamental form
3. Geometry of the Gauss Map
Definitions – Fundamental properties – In local coordinates
4. Intrinsic geoemtry of surfaces
Gauss theorem – Parallel transport – Gauss-Bonnet theorem
5. Other topics

1.3 Parametrized Curves

We denote by R3 the set of triples (x, y, z) of real numbers. Our goal is to characterize certain subsets of R3
(to be called curves) that are, in a certain sense, one-dimensional and to which the methods of differential
calculus can be applied.

Definition 1.1 A parametrized differentiable curve is a differentiable map α : I → R3 of an open interval

I = (a, b) of the real line R to R3 .

Remark 1.2 We say that a real function of a real variable is differentiable (or smooth) if it has, at all
points, derivatives of all orders (which are automatically continuous).

1-1
1-2 Lecture 1: Curves

Remark 1.3 The word differentiable in this definition means that α is a correspondence which maps each t ∈
I into a point α(t) = (x(t), y(t), z(t) ∈ R3 in such a way that the functions x(t), y(t), z(t) are differentiable.

The variable t is called the parameter of the curve. The word interval is taken in a generalized sense, so
that we do not exclude the cases a = −∞, b = ∞. The image set α(I) ⊂ R3 is called the trace of α.

Definition 1.4 The vector (x0 (t), y 0 (t), z 0 (t)) = α0 (t) ∈ R3 is called the tangent vector (or velocity vector)
of the curve α at t.

Definition 1.5 Let u = (u1 , u2 , u3 ) ∈ R3 and define its norm (or length) by
q
|u| = u21 + u22 + u23 .

Geometrically, |u| is the distance from the point (u1 , u2 , u3 ) to the origin O = (0, 0, 0).

Definition 1.6 (Inner product) Let u = (u1 , u2 , u3 ) ∈ R3 and v = (v1 , v2 , v3 ) ∈ R3 , and let θ, where
~ and Ov.
θ ∈ [0, π], be the angle formed by the segments Ou ~ The inner product u · v is defined by

u · v = |u||v| cos θ.

Proposition 1.7 The following properties hold:

1. Assume that u and v are nonzero vectors. Then u · v = 0 if and only if u is orthogonal to v.
2. u · v = v · u.
3. λ(u · v) = λu · v = u · λv.
4. u · (v + w) = u · v + u · w.

Proof: They are obvious by Definition 1.6. In fact, the fourth property is easy to see based on the observation
that the projection of v + w onto u is equal to the projection of v onto u plus the projection of w onto u.

Proposition 1.8 Let u = (u1 , u2 , u3 ) ∈ R3 and v = (v1 , v2 , v3 ) ∈ R3 , then

u · v = u1 v1 + u2 v2 + u3 v3 .

Proof: Let e1 = (1, 0, 0), e2 = (0, 1, 0), and e3 = (0, 0, 1). Then ei · ej = 1 if i = j, and ei · ej = 0 if i 6= j.
So u = u1 e1 + u2 e2 + u3 e3 and v = v1 e1 + v2 e2 + v3 e3 . By Proposition 1.7, we obtain

u · v = u1 v1 + u2 v2 + u3 v3 .

It easily follows from Proposition 1.8 that

Proposition 1.9 If u(t) and v(t), t ∈ I, are differentiable curve, then u(t) · v(t) is a differentiable function,
and
d
(u(t) · v(t)) = u0 (t) · v(t) + u(t) · v 0 (t).
dt
Lecture 1: Curves 1-3

Definition 1.10 A parametrized differential curve α : I → R3 is said to be regular if α0 (t) 6= 0 for all t ∈ I.

Recall that given t0 ∈ I, the arc length of a regular parametrized curve α : I → R3 , from point t0 , is given
by
Z t
s(t) = |α0 (t)|dt,
t0
p
where |α0 (t)| = (x0 (t))2 + (y 0 (t))2 + (z 0 (t))2 is the length of the vector α0 (t). Because α0 (t) 6= 0, the arc
length s is a differentiable function of t, ds/dt = |α0 (t)|, and dt/ds = 1/|α0 (t)|. Then t can be regarded as a
function of s, i.e., t = t(s). Let α(s) = α(t(s)). Then |α0 (s)| ≡ 1 and
Z s
s − s(t0 ) = |α0 (s)|ds.
s(t0 )

Definition 1.11 If |α0 (t)| ≡ 1 for all t ∈ I, then we call t is the arc length of α.

Remark 1.12 In this course, we will only consider regular parametrized differentiable curves parametrized
by arc length.

1.4 Vector Product in R3

If we express u, v, and w in the natural basis {ei }3i=1 ,

3
X 3
X 3
X
u= ui ei , v= vi e i , w= wi ei ,
i=1 i=1 i=1

then we define
u1 u2 u3
det(u, v, w) := v1 v2 v3 ,
w1 w2 w3
where |aij | denotes the determinant of the matrix (aij ).

Definition 1.13 (vector product) Let u, v ∈ R3 . The vector product of u and v (in that order) is the
unique vector u ∧ v ∈ R3 characterized by

(u ∧ v) · w = det(u, v, w) for all w ∈ R3 .

Choosing w = ei for i = 1, 2, and 3, then it is immediate from Definition 1.13 that

u2 u3 u u3 u u2
u∧v = e − 1 e + 1 e . (1.1)
v2 v3 1 v1 v3 2 v1 v2 3

Proposition 1.14 The following properties are hold:

1. u ∧ v = −v ∧ u (anti-commutativity).

2. (au + bw) ∧ v = au ∧ v + bw ∧ v for any real numbers a and b.

1-4 Lecture 1: Curves

3. u ∧ v = 0 if and only if u and v are linearly dependent.

4. (u ∧ v) · u = 0 and (u ∧ v) · v = 0.

Remark 1.15 It follows from property 4 that the vector product u ∧ v is normal to a plane generated by u
and v.

By Definition 1.13, we know that det(u, v, u ∧ v) = (u ∧ v) · (u ∧ v) = |u ∧ v|2 > 0 is positive. Therefore,

right-hand rule applies.
Next, we prove the relation for arbitrary vectors u, v, x, and y,
u·x v·x
(u ∧ v) · (x ∧ y) = . (1.2)
u·y v·y
This can easily be done by observing that both sides are linear in u, v, x, and y. Thus it suffices to check
that
e · e ej · ek
(ei ∧ ej ) · (ek ∧ el ) = i k
ei · el ej · el
for all i, j, k, l = 1, 2, 3. This is a straightforward verification. It follows that
u·u u·v
|u ∧ v|2 = (u ∧ v) · (u ∧ v) = = |u|2 |v|2 (1 − cos2 θ) = |u|2 |v|2 sin2 θ.
u·v v·v
It is
|u ∧ v| = |u||v| sin θ. (1.3)
Finally, it follows from (1.1) that
d
(u(t) ∧ v(t)) = u0 (t) ∧ v(t) + u(t) ∧ v 0 (t). (1.4)
dt

1.5 Local Theorey of Curves Parametrized by Arc Length

Let α : I = (a, b) → R3 be a curve parametrized by arc length s. Since the tangent vector α0 (s) has unit
length, the norm |α00 (s)| of the second derivative measures the rate of change of the angle which neighboring
tangents make with the tangent at s. Therefore, |α00 (s)| gives a measure of how rapidly the curve pulls away
from the tangent line at s. So we define

Definition 1.16 Let α : I → R3 be a curve parametrized by arc length s ∈ I. The number k(s) = |α00 (s)| is
called the curvature of α at s.

If α is a straight line, α(s) = us + v, where u and v are constant vectors (|u| = 1), then k ≡ 0. Conversely,
if k = |α00 (s)| = 0, then by integration α(s) = us + v, and the curve is a straight line.
At points where k(s) 6= 0, a unit vector n(t) in the direction α00 (s) is well defined by the equation
α00 (s) = k(s)n(s). (1.5)
In what follows, we shall restrict ourselves to curves parametrized by arc length at which k(s) 6= 0. In this
case, the inverse R = 1/k of the curvature is called the radius of curvature at s. Of course, a circle of radius
r has radius of curvature equal to r, as one can easily verify.
Moreover, α00 (s) is normal to α0 (s), because by differentiating α0 (s) · α0 (s) = 1 we obtain α00 (s) · α0 (s) = 0.
Lecture 1: Curves 1-5

Definition 1.17 n(s) defined by (1.5) is normal to α0 (s) and is called the normal vector at s. The plane
determined by the unit tangent and normal vectors, α0 (s) and n(s), is called the osculating plane at s.

Let
t(s) = α0 (s) and b(s) = t(s) ∧ n(s). (1.6)
Then
t0 (s) = k(s)n(s). (1.7)

Definition 1.18 b(s) defined in (1.6) is normal to the osculating plane and is called the binormal vector at
s.

Since b(s) is a unit vector, the length |b0 (s)| measures the rate of change of the neighboring osculating planes
with osculating plane at s, that is, |b0 (s)| measures how rapidly the curve pull away from the osculating
plane at s.
To compute b0 (s) we observe that, on the one hand, b0 (s) is normal to b(s), and that on the other hand,

b0 (s) = t0 (s) ∧ n(s) + t(s) ∧ n0 (s) = t(s) ∧ n0 (s).

It follows that b0 (s) is parallel to n(s), and we write

b0 (s) = τ (s)n(s) (1.8)

for some function τ (s).

Definition 1.19 Let α : I → R3 be a curve parametrized by arc length s such that α00 (s) 6= 0, s ∈ I. The
number τ (s) defined in (1.8) is called the torsion of α at s.

Remark 1.20 In constrast to the curvature, the torsion may be either positive or negative.

If α is a plane curve (i.e., α(I) is contained in a plane), then the plane of the curve agrees with the osculating
plane. Hence τ ≡ 0. Conversely, if τ ≡ 0 (and k 6= 0), we have that b(s) = b0 = constant, and therefore

(α(s) · b0 )0 = α0 (s) · b0 = 0.

It follows that α(s) · b0 = constant. Hence α(s) is contained in a plane normal to b0 .

Definition 1.21 Let α : I → R3 be a curve parametrized by arc length s such that α00 (s) 6= 0, s ∈ I. For
each s ∈ I, we have associated three orthogonal unit vectors t(s), n(s), b(s). The trihedron thus formed is
referred to as the Frenet trihedron at s.

Finally, note that it follows from (1.6) that n = b ∧ t. Then

n0 (s) = b0 (s) ∧ t(s) + b(s) ∧ t0 (s) = −τ (s)b(s) − k(s)t(s).

That is
n0 (s) = −k(s)t(s) − τ (s)b(s). (1.9)

Definition 1.22 We call equations (1.7), (1.8), and (1.9), the Frenet formulas.
1-6 Lecture 1: Curves

Lemma 1.23 Let α : I → R3 be a regular parametrized curve with arc length s. Then the curvature of α at
s ∈ I is
k(s) = |α0 ∧ α00 |,
and the torsion of α at s ∈ I is
(α0 ∧ α00 ) · α000
τ (s) = − .
|α0 ∧ α00 |2

Proof: By the straightforward calculation, we have

α0 (s) =t(s)
α00 (s) =k(s)n(s)
α000 (s) =k 0 (s)n(s) − k 2 (s)t(s) − k(s)τ (s)b(s).

Therefore,
α0 ∧ α00 = kt ∧ n = kb
and then
(α0 ∧ α00 ) · α000 = −k 2 τ.
Hence this lemma follows directly.
Physically, we can think of a curve in R3 as being obtained from a straight line by bending (curvature) and
twisting (torsion). After reflecting on this construction, we are led to conjecture the following statement,
which, roughly speaking, shows that k and τ describe completely the local behavior of the curve.

Theorem 1.24 (Fundamental theorem of the local theory of curves) Given differential functions k(s) >
0 and τ (s), s ∈ I, there exists a regular parametrized curve α : I → R3 such that s is the arc length, k(s)
is the curvature, and τ (s) is the torsion of α. Moreover, any other curve α̃, satisfying the same condi-
tions, differs from α by a rigid motion, that is, there exists an orthogonal linear map ρ of R3 , with positive
determinant, and a vector c such that α̃ = ρ ◦ α + c.

Proof: Observe that Frenet’s equations (1.7), (1.8), and (1.9) can be considered as a differential system in
I × R9 as 
dξ1
 ds = f1 (s, ξ1 , ξ2 , · · · , ξ9 )

··· , s ∈ I, (1.10)
 dξ9

ds = f9 (s, ξ1 , ξ2 , · · · , ξ9 )

where t = (ξ1 , ξ2 , ξ3 ), n = (ξ4 , ξ5 , ξ6 ), b = (ξ7 , ξ8 , ξ9 ), and fi , for i = 1, · · · , 9, are linear functions (with
coefficients that depend on s) of the coordinates ξi .
First, let us introduce the following theorem of existence and unqueness for such linear differential system
(cf. S. Lang, Analysis I, Addison-Wesley, Reading, Maxx., 1968, pp. 383-386):
Let fi be Lipschitz continuous. Given initial condition that at s0 ∈ I, (ξ1 , · · · , ξ9 ) = ((ξ1 )0 , · · · , (ξ9 )0 ), there
exists an open interval J ⊂ I containing s0 and a unique differential mapping α : J → R9 , with

α(s0 ) = ((ξ1 )0 , · · · , (ξ9 )0 ) and α0 (s) = (f1 , · · · , f9 ),

where each fi , i = 1, · · · , 9, is calculated in (s, α(s)) ∈ J × R9 . Furthermore, if the system is linear, J = I.

Therefore, it follows that given an orthonormal, positively oriented trihedron {t0 , n0 , b0 } in R3 and a value
s0 ∈ I, there exists a family of trihedrons {t(s), n(s), b(s)}, s ∈ I with t(s0 ) = t0 , n(s0 ) = n0 , b(s0 ) = b0 and
satisfying equations (1.7), (1.8), and (1.9).
Lecture 1: Curves 1-7

First, we will show {t(s), n(s), b(s)} remains orthonormal for every s ∈ I. In fact, by straightforward
computation, we know that the six quantities t · n, t · b, n · b, t · t, n · n, and b · b satisfy the following
differential equations: 

 (t · n)0 = k(n · n) − k(t · t) − τ (t · b)
(t · b)0 = k(n · b) + τ (t · n)





(n · b)0 = −k(t · b) − τ (b · b) + τ (n · n)




 (t · t)0 = 2k(t · n)
(n · n)0 = −2k(n · t) − 2τ (n · b)





(b · b)0 = 2τ (b · n)


It is easy to check that

t·n=t·b=n·b=0 and t·t=n·n=b·b=1

is a solution of the above system with initial conditions 0, 0, 0, 1, 1, 1. By uniqueness, it is the only solution,
so that the family {t(s), n(s), b(s)} is orthonormal for every s ∈ I.
Now, define Z
α(s) = t(s)ds, s ∈ I.

It is clear that α0 (s) = t(s) and α00 (s) = t0 (s) = kn. Therefore, k(s) is the curvature of α at s. Moreover,
since
α000 (s) = k 0 n + kn0 = k 0 n − k 2 t − kτ b,
and by Lemma 1.23, we know the torsion of α is given by

(α0 ∧ α00 ) · α000 (t ∧ (kn)) · (k 0 n − k 2 t − kτ b)

− 2
=− = τ.
k k2
Therefore, the torsion of α is τ .
We still have to show that α is unique up to translation and rotations of R3 .
We first remark that are length, curvature, and torsion are invariant under rigid motions. For instance, if
M : R3 → R3 is a rigid motion and α = α(t) is a parametrized curve, then
b b
d(M ◦ α)
Z Z
dα
dt = dt.
a dt a dt
A careful checking can be left as an exercise.
Now, assume that two curves α = α(s) and α̃ = α̃(s) satisfy the conditions k(s) = k̃(s) and τ (s) = τ̃ (s),
s ∈ I. Let t0 = t(s0 ), n0 = n(s0 ), b0 = b(s0 ), and t̃0 = t̃(s0 ), ñ0 = ñ(s0 ), b̃0 = b(s0 ) be the Frenet
trihedrons at s = s0 ∈ I of α and α̃, respectively. Clearly, there is a rigid motion which takes α̃(s0 ) into
α(s0 ) (translation) and t̃0 , ñ0 , and b̃0 into t0 , n0 , and b0 (rotation). Thus, after performing this rigid motion
on α̃, we have that α̃(s0 ) = α(s0 ) and that the Frenet trihedrons t(s), n(s), b(s), and t̃(s), ñ(s), b̃(s), of α
and α̃, respectively, satisfy the Frenet equations:
 
dt dt̃
 ds = kn
  ds = kñ

dn dñ
ds = −kt − τ b ds = −k t̃ − τ b̃
 db
 
 db̃
ds = τ n ds = τ ñ

with t(s0 ) = t̃(s0 ), n(s0 ) = ñ(s0 ), and b(s0 ) = b̃(s0 ). By applying the uniqueness part of the above theorem
on differential equations, we obtain the desired result.