Linear First-Order PDEs

MA 201: Partial Differential Equations

Lecture - 2

MA201 (2022): PDE

 Linear First-Order PDEs

How do first-order PDEs occur?

• First-order PDEs mainly connect to geometry.
• Two-parameter family of surfaces: Let

f (x, y , u, a, b) = 0 (1)

represent two-parameter family of surfaces in R3 , where a and b are

arbitrary constants.
Differentiating (1) with respect to x and y yields the relations

∂f ∂f
+p = 0, (2)
∂x ∂u
∂f ∂f
+q = 0. (3)
∂y ∂u

Eliminating a and b from (1), (2) and (3), we get a relation of the
form
F (x, y , u, p, q) = 0, (4)
which is a first-order PDE for the unknown function u of two
independent variables x and y .
MA201 (2022): PDE
 Linear First-Order PDEs

Example
The equation
x 2 + y 2 + (u − c)2 = r 2 , (5)
where r and c are arbitrary constants, represents the set of all spheres
whose centers lie on the u-axis.
Differentiating (5) with respect to x, we obtain

∂u
x + (u − c) = 0. (6)
∂x
Differentiate (5) with respect to y to have

∂u
y + (u − c) = 0. (7)
∂y

Eliminating the arbitrary constant c from (6) and (7), we obtain a

first-order PDE:
∂u ∂u
y −x =0 -linear PDE. (8)
∂x ∂y

MA201 (2022): PDE

 Linear First-Order PDEs

Example
The equation
x 2 + y 2 = (u − c)2 tan2 α, (9)
where c and α are arbitrary constants, represents a family of all right
circular cones having the u-axis as their axes.
Differentiating (9) with respect to x, we obtain
∂u
(u − c) tan2 α = x. (10)
∂x
Differentiate (9) with respect to y to have
∂u
(u − c) tan2 α = y . (11)
∂y

Eliminating the arbitrary constants c and α from (10) and (11), we

obtain a first-order PDE:
∂u ∂u
y −x = 0, (12)
∂x ∂y

which is interestingly the same as (8).

MA201 (2022): PDE
 Linear First-Order PDEs

• Unknown function of known functions

• Unknown function of a single known function
Let
u = f (g ), (13)
where f is an unknown function and g is a known function of two
independent variables x and y .
Differentiating (13) with respect to x and y , respectively, yields the
equations
ux = f ′ (g )gx (14)
and
uy = f ′ (g )gy , (15)
′ df
where f (g ) = .dg
Eliminating f ′ (g ) from (18) and (19), we obtain

gy ux − gx uy = 0,

which is a first-order PDE for u.

MA201 (2022): PDE

 Linear First-Order PDEs

Example
Consider the surfaces described by an equation of the form

u = f (x 2 + y 2 ), (16)

where f is an arbitrary function of a known function g (x, y ) = x 2 + y 2 .

Differentiating (16) with respect to x and y , it follows that

ux = 2xf ′ (g ), uy = 2yf ′ (g ),
df
where f ′ (g ) = dg .
′
Eliminating f (g ) from the above two equations, we obtain a first-order
PDE
yux − xuy = 0. (17)

It is clear that (17) is the same PDE as (8) and (12). This is because the
forms of (5) and (9) allow for both equations to possess the similar kind
of solutions. Both (5) and (9) have the similar properties as mentioned,
i.e., u = f (x 2 + y 2 ).

MA201 (2022): PDE

 Linear First-Order PDEs

• Unknown functions of two known functions

Let
u = f (x − ay ) + g (x + ay ), (18)
where a > 0 is a constant.
With v (x, y ) = x − ay and w (x, y ) = x + ay , we can write (18) as

u = f (v ) + g (w ). (19)

Differentiating (19) w. r. t. x and y , respectively, yields

p = ux = f ′ (x − ay ) + g ′ (x + ay ),
q = uy = −af ′ (x − ay ) + ag ′ (x + ay ).

Eliminating f ′ (v ) and g ′ (w ), we get

qy = a 2 px .

In terms of u, the above PDE is the well-known one-dimensional

wave equation
uyy = a2 uxx .
MA201 (2022): PDE
 Linear First-Order PDEs

Example (Geometrical problem)

All functions u(x, y ) such that the tangent plane to the graph
u = u(x, y ) at any arbitrary point (x0 , y0 , u(x0 , y0 )) passes through the
origin is characterized by the PDE xux + yuy − u = 0.
The equation of the tangent plane at (x0 , y0 , u(x0 , y0 )) is

ux (x0 , y0 )(x − x0 ) + uy (x0 , y0 )(y − y0 ) − (u − u(x0 , y0 )) = 0.

Since this plane passes through the origin (0, 0, 0), we have

−ux (x0 , y0 )x0 − uy (x0 , y0 )y0 + u(x0 , y0 ) = 0. (20)

For equation (20) to hold for all (x0 , y0 ) in the domain of u, it must be
that u satisfies
xux + yuy − u = 0,
which is a first-order linear PDE.

MA201 (2022): PDE

 Linear First-Order PDEs

Cauchy’s problem or IVP for first-order PDEs:

Let Γ be a given curve in R2 described parametrically by the equations
x = x0 (s), y = y0 (s); s ∈ I, (21)
where x0 (s), y0 (s) are in C 1 (I ).
The IVP or Cauchy’s problem for first-order PDE
F (x, y , u, p, q) = 0 (22)
is to find a function φ = φ(x, y ) with the following properties:
• φ(x, y ) and its partial derivatives with respect to x and y are
continuous in a region Ω of R2 containing the curve Γ.
• φ = φ(x, y ) is a solution of (22) in Ω, i.e.,

F (x, y , φ(x, y ), φx (x, y ), φy (x, y )) = 0 in Ω.

• On the curve Γ, we have

φ(x0 (s), y0 (s)) = u0 (s), s ∈ I . (23)

The curve Γ is called the initial curve of the problem and the function
u0 (s) is called the initial data. Equation (23) is called the initial
condition (or side condition) of the problem.
MA201 (2022): PDE
 Linear First-Order PDEs

• Well-posed Problem (In the sense of Hadamard)

The Cauchy’s problem (PDE + side condition) is said to be well-posed if
it satisfies the following criteria:
1 The solution exists.
2 The solution is unique.
3 The solution depends continuously on the initial and/or boundary
data.
If one or more of the above conditions does not/do not hold, we say that
the problem is ill-posed.

MA201 (2022): PDE

 Linear First-Order PDEs

Linear First-Order PDEs

The most general first-order linear PDE has the form

a(x, y )ux + b(x, y )uy = c(x, y )u + d(x, y ), (24)

where a, b, c, and d are given functions of x and y . These functions are
assumed to be continuously differentiable.
Observe the left hand side of (24):

a(x, y )ux + b(x, y )uy = ∇u · (a, b),

which is (essentially) a directional derivative of u(x, y ) in the direction of

the vector (a, b), where (a, b) is defined and is non-zero.
Remark: When a and b are constants, the vector (a, b) has a fixed
direction and magnitude, but now it is seen that the vector (a, b) can
change as its base point (x, y ) varies. Thus, (a, b) is a vector field on the
plane.

MA201 (2022): PDE

 Linear First-Order PDEs

The equations
dx dy
= a(x, y ), = b(x, y ), (25)
dt dt
determine a family of curves x = x(t), y = y (t) whose tangent vector
dy
( dx
dt , dt ) coincides with the direction of the vector (a, b). Here t is
considered as a parameter, not time as usually considered.
Therefore,
d ∂u dx ∂u dy
u{(x(t), y (t))} = +
dt ∂x dt ∂y dt
= ux (x(t), y (t))a(x(t), y (t))
+uy (x(t), y (t))b(x(t), y (t))
= c(x(t), y (t))u(x(t), y (t)) + d(x(t), y (t))
= c(t)u(t) + d(t),

where we have used the chain rule and (24).

Thus, along these curves, u(t) = u(x(t), y (t)) satisfies the ODE

u ′ (t) − c(t)u(t) = d(t). (26)

MA201 (2022): PDE

 Linear First-Order PDEs
h R i
t
Let µ(t) = exp − 0 c(τ )dτ be an integrating factor for (26). Then,
the solution is given by
Z t 
1
u(t) = µ(τ )d(τ )dτ + u(0) . (27)
µ(t) 0

The approach described above is called the method of characteristics. It

is based on the geometric interpretation of the partial differential
equation (24).
Remarks.
• The system of ODEs (25) is known as the characteristic equation for
the PDE (24). The solution curves of the characteristic equation are
the characteristic curves for (24).
• The values u(t) of the solution u along the entire characteristics
curve are completely determined, once the value
u(0) = u(x(0), y (0)) is prescribed.
• Assuming certain smoothness conditions on the functions a, b, c,
and d, the existence and uniqueness theory for ODEs guarantees a
unique solution curve (x(t), y (t), u(t)) of (25) and (26).

MA201 (2022): PDE