DISCLAIMER: These lecture notes consist of a compilation of material

taken from numerous sources, and so should not be taken as original text.
Note that there could be some typos in the text.

1
MAT 5122-PARTIAL DIFFERENTIAL EQUATIONS (PDEs)
TOPIC 2: First Order Partial
differential Equations

0.1 Introduction

A first order PDE for an unknown function u(x1 , x2 , · · · , xn ) has the following form:
F (x1 , x2 , · · · , xn , u, ux1 , ux2 , · · · , uxn ) = 0 (1)
where F is a given function of (2n + 1) variables. Many problems in the mathematical,
physical and engineering sciences deal with the formulation and solution of first-order PDEs.
In applications, first-order PDEs are most commonly used to describe dynamical processes,
and so time, t, is one of the independent variables. First-order PDEs and systems model
a wide variety of wave phenomena, including transport of solvents in fluids, flood waves,
acoustics, gas dynamics, glacier motion, traffic flow, and also a variety of biological and
ecological systems. For simplicity in this topic we shall concentrate on functions in two
variables.

Consider a surface in R3 whose graph is given by u(x, y). The surface satisfies an equation
of the form
F (x, y, u, ux , uy ) = 0. (2)
Since u(x, y) is a surface in R3 , and since the normal to the surface is given by the vector
hux , uy , −1i, the PDE (2) can be considered as an equation relating the surface to its normal.

If equation (2) can be written in the form

a(x, y, u)ux + b(x, y, u)uy = c(x, y, u) (3)
then we say that the equation is quasi-linear. If it can be written in the form
a(x, y)ux + b(x, y)uy = c(x, y, u) (4)

1
then we say that the equation is semi-linear. Further, if equation (2) can be written in the
form
a(x, y)ux + b(x, y)uy + c(x, y)u = d(x, y) (5)
then we say that the equation is linear. A first-order PDE that is not linear is said to be
non-linear.

A linear first-order PDE is called homogeneous if d(x, y) = 0 and non-homogeneous if

d(x, y) 6= 0.

0.2 Quasi-linear PDEs

The general form is (3)

Example 1. Solve the IVP ux = c0 u + c1 , u(0, y) = y, where c0 is a constant and c1 is a
function of x and y.

Solution: ux − c0 u = c1 (x, y) ⇒ u(x, y)e−c0 x = e−c0 x c1 (x, y)dx + A(y)

R
Z Z x
−c0 x
⇒ u(x, y) = e c0 x
e c0 x
c1 (x, y)dx + e A(y) ⇒ u(x, y) = e c0 x
e−c0 t c1 (t, y)dt + ec0 x A(y)
0

u(0, y) = A(y) ⇒ A(y) = y

Z x
⇒ u(x, y) = e c0 x
e−c0 t c1 (t, y)dt + ec0 x y.
0

Method of characteristics

The method consists of transforming the PDE to a system of ODEs which can be solved and
the found solution is transformed into a solution for the original PDE. The method relies
on a geometrical argument. The solution is visualised as an integral surface with equation
z = u(x, y). An alternative representation of this integral surface is:
F (x, y, z) = u(x, y) − z = 0.
That is, an integral surface is a level surface of the function F (x, y, z). The normal vector to
the surface F (x, y, z) = 0 is given by
n̂ = OF = Fx i + Fy j + Fz k = ux i + uy j + uz k.

2
 Figure 1: Solution Surface

Equation (3) can be written as the dot product (see figure 2)

ha(x, y, u), b(x, y, u), c(x, y, u)i.hux , uy , −1i = 0. ∗∗
To solve (3) amounts to constructing a surface such that the normal to the surface sat-
isfys the orthogonality condition (**). This is equivalent to saying that we seek a sur-
face u = u(x, y) such that the tangent plane to the surface at (x, y, u) contains the vector
ha(x, y, u), b(x, y, u), c(x, y, u)i.

We seek a solution of (3) that contains a given curve, say

Γ = Γ(s) = hx0 (s), y0 (s), u0 (s)i, s ∈ I = (α, β). (6)
The curve Γ is called the initial curve. Recall that a three dimensional surface (in our
case the solution surface u = u(x, y)) can be represented parametrically as a two pa-
rameter family hx(s, t), y(s, t), u(s, t)i. In order that the initial values be achieved we need
hx(s, 0), y(s, 0), u(s, 0)i = hx0 (s), y0 (s), u0 (s)i. If these curves lie on a solution surface, then
tangent vectors to the curves must lie in the tangent plane to the surface at the point. This
means that for a fixed s the curves (in t) must satisfy the equations
dx
(s, t) = a(x, y, u), x(s, 0) = x0 (s), (7)
dt
dy
(s, t) = b(x, y, u), y(s, 0) = y0 (s),
dt
du
(s, t) = c(x, y, u), u(s, 0) = u0 (s).
dt

3
 Figure 2: Solution Surface and Characteristic

If the vector field A = ha(x, y, u), b(x, y, u), c(x, y, u)i were tangent to the curve Γ, then a
solution of 7 would coincide with Γ and our method for constructing a surface would fail.
This problem will be avoided by not prescribing data on the curve with
dx dy
= a, = b.
dt dt
The system
dx
(s, t) = a(x, y, u),
dt
dy
(s, t) = b(x, y, u),
dt
du
(s, t) = c(x, y, u)
dt

is called a system of characteristic equations, or in short, the characteristic equations. The

solutions are called characteristic curves of the equation. Each point on the initial curve Γ
is a starting point for a characteristic curve. The system 7 is called the Cauchy problem
for quasi-linear equations.

To construct a solution surface we first find

x = x(s, t), y = y(s, t), u = u(s, t)

4
and then solve the two equations x = x(s, t), and y = y(s, t) for s and t in terms of x and
y. This is an inversion of the transformation from (t, s) plane to the (x, y) plane. Recall that
the implicit function theorem implies that such a transformation is invertible if the Jacobian
 ∂x
(s0 , t0 ) ∂x

∂(x, y) (s0 , t0 )
J= ∂t
= det ∂y ∂s 6= 0.
∂(t, s) ∂t
(s0 , t0 ) ∂y (s , t )
∂s 0 0

An explicit computation of the Jacobian at points located on the initial curve Γ, using the
characteristic equations, gives
 
∂x ∂y ∂x ∂y  a b 
J= − = = (y0 )s a − (x0 )s b. (8)
∂t ∂s ∂s ∂t (x0 )s (y0 )s 
Thus the Jacobian vanishes at some point if and only if the vectors ha, bi and h(x0 )s , (y0 )s i
are linearly dependent. Hence the geometrical meaning of a vanishing Jacobian is that
the projection of Γ on the (x, y) plane is tangent at this point to the projection of the
characteristic curve on that plane. Hence, in order for a first-order quasilinear PDE to have
a unique solution near the initial curve, we must have J 6= 0. This condition is called the
transversality condition.
Theorem 2. (Existence and Uniqueness of linear and quailinear PDEs) Assume that the
coefficients of the quasilinear equation (3) are smooth functions of their variables in a neigh-
bourhood of the initial curve. Assume further that the transversality condition holds at each
point s in the interval s0 − 2δ, s0 + 2δ on the initial curve. Then the Cauchy problem (7) has
a unique solution in the neighbourhood (t, s) ∈ (−, ) × (s0 − δ, s0 = δ) of the initial curve.

If the transversality condition does not hold for an interval of s values, then the Cauchy
problem has either no solution at all, or it has infinitely many solutions.
Example 3. Solve each of the following problems.

1. ux + uy = 2; u(x, 0) = x2 .
Solution: Here a(x, y, u) = 1, b(x, y, u) = 1, u(x, y, u) = 2. Hence we have three
systems of first order problems to be solved:
2
xt (t, s) = 1, ; x(0, s) = s : yt (t,  s) = 1;  y(0, s) = 0 : ut (t, s) = 2; u(0, s) = s .
 a b  1 1
The Jacobian is J =  = = −1 6= 0, and the coefficients are smooth
(x0 )s (y0 )s  1 0
functions. Therefore, we expect to have a unique solution in the neighbourhood of the
initial curve hs, 0, s2 i. Next we solve the problems.
x(t, s) = t + f1 (s x(0, s) = f1 (s) = s and so x(t, s) = t + s.
y(t, s) = t + f2 (s), y(0, s) = f2 (s) = 0 and so y(t, s) = t.

5
 u(t, s) = 2t + f3 (s), u(0, s) = f3 (s) = s2 , and so, u(t, s) = 2t + s2 .
So the parametric representation of the solution surface is:

x = t + s, y = t, u = 2t + s2 .
The explicit representation of the solution surface u as a function of x and y is u(x, y) =
2y + (x − y)2 .
2. ux = 1; u(0, y) = g(y).
Solution: xt (t, s) = 1; x(0, s) = 0 ⇒ x(t, s) = t + h1 (s). x(0, s) = h1 (s) = 0 and
so x(t, s) = t
yt (t, s) = 0; y(0, s) = s ⇒ y(t, s) = h2 (s). y(0, s) = h2 (s) = s and so y(t, s) = s
ut (t, s) = 1; u(0, s) = g(s) ⇒ u(t, s) = t + h3 (s). u(0, s) = h3 (s) = g(s) and
so u(t, s) = t + g(s). Therefore we have parametric surface ht, s, t + g(s)i, and hence
u(x, y) = x + g(y).
3. ux + uy + u = 1; u = sin x on y = x + x2 , x > 0.
Solution: xt (t, s) = 1; x(0, s) = s ⇒ x(t, s) = t + g1 (s). x(0, s) = g1 (s) = s, and
so x(t, s) = t + s.
yt (t, s) = 1; y(0, s) = s + s2 ⇒ y(t, s) = t + g2 (s). y(0, s) = g2 (s) = s + s2 , and so
y(t, s) = t + s + s2 .
∂
ut (t, s) + u = 1; u(0, s) = sin s ⇒ ∂t (uet ) = et ⇒ et u(t, s) = et + g3 (s)
u(t, s) = 1 + e−t g3 (s). u(0, s) = 1 + g3 (s) = sin s ⇒ g3 (s) = sin s − 1, and so
u(t, s) = 1 + e−t (sin s − 1). Parametrically, the solution surface is:

x = t + s, y = t + s + s2 , u = 1 + e−t (sin s − 1).

√ √
Therefore, u(x, y) = 1 + (sin( y − x) − 1)e y−x−x .
4. −yux + xuy = u; u(x, 0) = ϕ(x)
Solution xt (t, s) = −y and yt (t, s) = x; x(0, s) = s, y(0, s) = 0, u(0, s) = ϕ(s).
dy
dx
= − xy ⇒ y 2 + x2 = c and
y 2 (0, s) = c − x( 0, s) ⇒ 0 = c − s2 ⇒ c = s2 and so y 2 + x2 = ss
ut (t, s) = u; u(0, s) = ϕ(s) ⇒ u(t, s) = ϕ(s)et . Parametrically the integral surface is
x(t, s) = s cos t, y(t, s) = s sin t, u(t, s) = et ϕ(s).
Therefore, p y
u(x, y) = ϕ( x2 + y 2 )earctan( x ) .

6
Bibliography

[1] Yeduda Pnchover and Jacob Rubinstein, An Introduction to Partial Differential Equa-
tions, New York, 2005.
[2] Ravi P. Agarwal and Donald O’Regan, Differential Equations, with Special Functions,
Fourier Series and Boundary Value Problems, Springer Science + Business Media,
2009.
[3] Jeffrey R. Chasnov, Introduction to Differential Equations. Lecture notes for MATH
2351/2352 , The Hong Kong University of Science and Technology, Department of
Mathematics.
[4] Marcel B. Finan, Lecture Notes in Mathematics Arkansas Tech University, Department
of Mathematics, 2018.