You are on page 1of 70

ASSIGNMENT

ON

Qualitative Theory

COURSE TITLE: Population dynamics

COURSE NO: MATH-5209

SUBMITTED TO SUBMITTED BY
Dr. Md. Haider Ali Biswas Tahera Parvin
Professor Student ID: M.Sc- 201203
MATHEMATICS DISCIPLINE
MATHEMATICS DISCIPLINE KHULNA UNIVERSITY, KHULNA
KHULNA UNIVERSITY, KHULNA

SUBMISSION DATE: 05.10.2020

MATHEMATICS DISCIPLINE, KHULNA UNIVERSITY, KHULNA


Nonlinear Systems of Differential Equations—Consumer-Resource Models

Nonlinear, autonomous systems of ordinary differential equations are of the form

 f1 x1 , x 2 ,, x n 
dx1
dt
 f 2 x1 , x 2 ,, x n 
dx2
dt

dxn
 f n x1 , x 2 ,, x n 
dt

Where each of the functions fi on the right-hand side are real-valued functions in n variables. Most of
the time, we will restrict the analysis to systems of two variables. We will focus on Equilibria and
stability.

Equilibria and Stability

Consider the system of two autonomous differential equations

dx
 f ( x, y )
dt
(1)
dy
 g ( x, y )
dt

The first step is to find the equations of the zero isoclines, which are defined as the set of points that
satisfy

0  f ( x, y )
0  g ( x, y )
Each equation results in a curve in the x-y space. Point Equilibria occur where the two isoclines
intersect (Figure 1). A point equilibrium ( xˆ , yˆ ) of (1) therefore simultaneously satisfies the two
equations

f ( xˆ, yˆ )  0 and g ( xˆ, yˆ )  0

We will call point Equilibria simply “Equilibria.”


y
f ( x, y)  0

Equilibrium

g ( x, y)  0

Figure 1: Zero isoclines corresponding to the two differential equations. Equilibria occur where the
isoclines intersect.

The analytical approach to stability relies on analyzing the effects of small perturbations. We say that
the equilibrium ( xˆ , yˆ ) is locally stable if the system returns to ( xˆ , yˆ ) after a small perturbation, and
unstable otherwise. Mathematically, this can be analyzed through linearizing the right-hand side of
each of the two differential equations in (1) about the equilibrium.
The system (1) can be written in matrix form

d  x(t )   f ( x, y )

dt  y (t )  g ( x, y ) 

where the right-hand side is a vector-valued function that maps a point in R2 (the two-dimensional real
plane) into a point in R2. To linearize a vector-valued function, we need to linearize each component
separately. Linearizing a function of two variables about a specific point means to find the tangent
plane at this point (this, of course, is not always possible). The equation of a tangent plane of f(x,y)
about ( xˆ , yˆ ) is given by

f ( xˆ, yˆ ) f ( xˆ, yˆ )
 ( x, y )  f ( xˆ, yˆ )  ( x  xˆ )  ( y  yˆ )
x y

 f ( x, y ) 
We thus find for the linearization of the vector-valued function  
 g ( x, y ) 

 f ( xˆ, yˆ ) f ( xˆ, yˆ ) 
 ( x, y )   f ( xˆ, yˆ )   x y   x  xˆ 
  ( x, y )    g ( xˆ, yˆ )    g ( xˆ, yˆ ) 
g ( xˆ, yˆ )   y  yˆ 
    
 x y 

Now, considering   x  xˆ and   y  yˆ as the perturbations, then with f ( xˆ, yˆ )  0 and g ( xˆ, yˆ )  0
, we find

 d   f ( xˆ , yˆ ) f ( xˆ, yˆ ) 
 dt   x y   
(2)  d    g ( xˆ, yˆ ) 
g ( xˆ, yˆ )   
  
 dt   x y 

The matrix

 f ( x, y ) f ( x, y ) 
 x y 
J ( x, y )   
 g ( x, y ) g ( x, y ) 
 x y 

is called the Jacobi matrix.

The system (2) is a linear system of two equations, and we can use the results from linear systems of
two differential equations to determine the stability of the Equilibria. Namely,
 The equilibrium is a node if both eigenvalues of the Jacobian evaluated at the equilibrium
are real, distinct, nonzero, and are of the same sign. The node is locally stable if the
eigenvalues are negative and unstable if the eigenvalues are positive.
 The equilibrium is a saddle if both eigenvalues of the Jacobian evaluated at the equilibrium
are real and nonzero but have opposite signs. A saddle is unstable.
 The equilibrium is a spiral if both eigenvalues of the Jacobian evaluated at the equilibrium
are complex conjugates with nonzero real parts. The spiral is locally stable if the real parts
of the eigenvalues are negative, and unstable if the real parts of the eigenvalues are
positive.

Euler Method for Systems of Differential Equations

To numerically solve a system of differential equations, we can use the Euler method. This is the
simplest approximation scheme. It has disadvantages, like slow convergence and instabilities. There
are far superior numerical schemes (like the Runge-Kutta method), but the Euler scheme suffices for
our purposes here.
The Euler scheme is directly based on the definition of the derivative. Namely,

dx x(t  h)  x(t )
 lim
dt h0 h

To approximate the system of differential equations


dx
 f ( x, y )
dt
dy
 g ( x, y )
dt

We approximate the left-hand side using the definition of the derivative

x(t  h)  x(t )
 f x(t ), y (t ) 
h
y (t  h)  y (t )
 g x(t ), y (t ) 
h

And then rewrite this as

x(t  h)  x(t )  hf x(t ), y (t ) 


y (t  h)  y (t )  hg x(t ), y (t ) 

Which can be coded up in a spreadsheet. The time step h needs to be small enough so that the
approximation is sufficiently good but not too small so that rounding errors accumulate too quickly.

Consumer-Resource Models

Model 1: Lotka-Volterra Predator-Prey Model

There are a large number of models that deal with consumer and resource interactions. The oldest such
model is the Lotka-Volterra model, which describes the interaction between a predator and its prey. If
x denotes the abundance of the prey and y the abundance of the predator, then the model is given by
the following set of differential equations:

dx
 bx  cxy
dt
(3)
dP
 ecxy  dy
dt

where the parameters b, c, e, and d on the right-hand side are positive constants. The parameter b is the
prey birth rate, c is the consumption rate, e is the conversion fraction of prey into new predators, and d
is the predator death rate. The Equilibria can be obtained algebraically by setting the right-hand sides
equal to 0 and solving for x and y. We find the trivial equilibrium (0,0) and a nontrivial equilibrium
xˆ, yˆ    d , b  . The corresponding Jacobi matrix is given by
 ec c 
b  cy  cx 
J ( x, y )   
 ecy ecx  d 

If we evaluate this at the trivial equilibrium (0,0), we find

b 0 
J (0,0)   
0  d 

Therefore, the two eigenvalues are 1  b and  2   d . Both eigenvalues are real. Since one
eigenvalue is positive and the other is negative, we conclude that the trivial equilibrium is a saddle and
thus unstable.
 d b
If we evaluate the Jacobi matrix at the nontrivial equilibrium xˆ, yˆ    ,  , we find
 ec c 

 d
0 c  d
 d b   ec    0  
J ,     e
 ec c  ec b 0  eb 0 
 c 

To compute the eigenvalues, we solve

  d b   d 
0    d
det  J  ,   I   det  e  0   0     eb      bd  0
2
  ec c    eb 0     e
 

Which yields the two complex conjugate eigenvalues 1  i bd and 2  i bd . The real parts of
these two eigenvalues are both equal to 0. The linear stability analysis is thus inconclusive.
It turns out that the equilibrium is neutral and this system of equations exhibits neutral
oscillations. This will be explored numerically in the computer lab exercises.

Model 2: Density-dependent Growth of the Prey

It is possible to stabilize the predator-prey model by including density-dependent growth of the prey in
the form of logistic growth. This takes the form

dx  x
 bx1    cxy
(4)
dt  K
dy
 ecxy  dy
dt
Where all parameters are positive. In the absence of the predator, the prey dynamics reduce to logistic
dx
growth in Equation (4). Namely, If we set y  0 , then  bx 1  x / K  .
dt

The system of equations (4) has the nontrivial equilibrium (0,0), which is always unstable. In addition,
d d
it has the prey only equilibrium (K,0), which is locally stable provided K  . If K  , an
ec ec
d
additional nontrivial equilibrium in the first quadrant appears, which is locally stable. If K  , the
ec
prey only equilibrium is unstable.

Model 3: Saturating Functional Response

The functional response of a predator is defined as the per predator predation rate. In this case, the
functional response is simply cx, a linear function of prey density. In nature, one often observes
functional response curves that saturate with increasing prey density. This happens, for instance, if
predators need to spend time handling prey.

A common form for a saturating functional response is

dx  x cx
 bx1    y
(5)
dt  K ax
dy cx
e y  dy
dt ax

This model is known as Rosenzweig’s model. It has the trivial equilibrium (0,0) and the prey only
ad
equilibrium (K,0). If 0   K , there is an additional equilibrium in the first quadrant, which
ec  d
may be locally stable or unstable. The interesting property of this model is that when increasing the
carrying capacity K, the equilibrium becomes unstable. This is known as Rosenzweig‟s paradox of
enrichment.

Nonlinear state models

 We will deal with the stability analysis of nonlinear systems, with emphasis on Lyapunov method.

 Consider dynamical systems that are modeled by a finite number of coupled first-order ordinary
differential equations as


x  f  t , x, u  ............................ 1

We call (1) the state equation and refer to x as the state and u as the input.
 Consider another equation

y  f  t , x, u  ................................  2 

Equation (2) is called the output.

State-space model

 Equations (1) and (2) are called together the state space model or simply state model.

Unforced State equation

 If the state model (1) is taken as


x  f  t , x  ..................... 3 

Where the input is specified as a function of time and state u    t , x  , then (3) is called unforced
state equation.

Implicit and explicit functions

In mathematics,

The explicit function is a function in which the dependent variable has been given “explicitly” in
terms of the independent variable. Or it is a function in which the dependent variable is expressed in
terms of some independent variables.

It is denoted by:

y=f(x)

Examples of Explicit functions are:

y=axn+bx where a, n and b are constant.

y=5x3-3

The Implicit function is a function in which the dependent variable has not been given “explicitly” in
terms of the independent variable. Or it is a function in which the dependent variable is not expressed
in terms of some independent variables.

It is denoted by:

R(x,y) = 0
Some examples of Implicit Functions are:
x2 + y2 – 1 = 0

Trajectory

In a dynamic system, a trajectory is the set of points in state space that are the future spaces resulting
from a given state.

Orthogonal Trajectories

Let

F ( x, y, e)  0 (1)

Let a given one parameter family of curses in the xy plane.

A curve that intersects the curves of the family (1) at right angle is called an orthogonal trajectory of
the given family.

For example: consider the family of circles.

x2  y 2  c2 (2)

with Centre at the origin and radius c.

Each straight line through the origin y  kx is an orthogonal trajectory of the family of circle (2).
y

x
Phase plane:

We consider a second order non linier differential equations of the form

d 2x  dx 
 F  x,  .......................... 1
 dt 
2
dt

As a specific example of such an equation we list the important Van-der Pol equation

d 2x
dt 2 
  x2 1
dx
dt

 x  0..................  2 

Where  is positive constant. We shall consider this equation at a later stage of our study. For the
time being , we merely note that (2) may be put in the form (1) where

 dx 

F  x,     x 2  1
 dt 
dx
dt
x
Let us suppose that the differential equation (1) describes a certain dynamical system having one
degree of freedom. The state of this system at time t is determined by the values of x (position) and
dx dx
(velocity). The plane of the variables x and is called a phase plane.
dt dt

Phase Portrait:
A phase portrait is a geometric representation of the trajectories of a dynamical system in the phase
plane. Each set of initial conditions is represented by a different curve, or point.
Phase portraits are an invaluable tool in studying dynamical systems. They consist of a plot of typical
trajectories in the state space. This reveals information such as whether an attractor, a repellor or limit
cycle is present for the chosen parameter value. The concept of topological equivalence is important in
classifying the behavior of systems by specifying when two different phase portraits represent the
same qualitative dynamic behavior. An attractor is a stable point which is also called 'sink'. The
repellor is considered as an unstable point, which is also known as 'source'.
A phase portrait graph of a dynamical system depicts the system's trajectories (with arrows) and stable
steady states (with dots) and unstable steady states (with circles) in a state space. The axes are of state
variables.

Critical Point:

Given the autonomous system


dx
 P  x, y 
dt (5)
dy
 Q  x, y 
dt

A point  x0 , y0  at which both P  x 0 , y0   0 and Q  x0 , y0   0

Is called a critical point of (5).

Isolated critical Point:

A critical point  x 0 , y0  of the system (5) is called isolated if there exist a circle

 x  x0    y  y0 
2 2
 r2

About the point  x 0 , y0  such that  x 0 , y0  is the only critical point of (5) with in this circle.

Autonomous system

In mathematics, an autonomous system or autonomous differential equation is a system of ordinary


differential equations which does not explicitly depend on the independent variable. When the variable
is time, they are also called time-invariant systems.
An autonomous system is a system of ordinary differential equations of the form

x  f  x

If the system is not autonomous, then it is called non-autonomous or time varying.

Equilibrium point

A point x  x in the state space is said to be an equilibrium point of x  f  x, t if f  x   0 . It has
the property that whenever the state of the system starts at x  x . It will remain at x  x of all future
time.

Mathematically,

x  t0   x  x  t   x , t  t0

Example:
Let the model ( ) 3 (1 )
Put ( ) , then 3 (1 )

3 3

3 2 0

(3 2) 0

0or 0.

Graphically,

For the autonomous systems, equilibrium points are the real roots of the equation f ( x)  0

Stable:
An equilibrium point x=0 of the state space is said to be stable if nearby solutions stay nearby for all
future time

Mathematically,

The equilibrium point x=0 of x  f  x  is stable if for each  0 , there is   0 such that

x  0  <   x  t  <  , t  0.

The equilibrium point x=0 is unstable if it is not stable.


Asymtotically stable:
Asymtotically stable if it is stable and   0 can be chosen such that

x  0  <   lim x t   0
t 

Theorem 1:

Suppose that f is a C1 function and x  x* in an equilibrium point of x  f ( x) , that is f ( x* )  0 .


Suppose also that f ( x* )  0.

Then the equilibrium point x* is asymptotically stable if f ( x)  0, and unstable if f ( x)  0.

Proof:

First, we prove stability in the sense of Lyapunov. Suppose   0 is given. We need to find a   0
such that for all x(0)   , it follows that x(t )   , t  0 . Let 1  min( , r ) . Define,

m  min f ( x)
x  1

Since f ( x) is a continuous, the above m is well defined and positive. Choose  satisfying 0    1
such that for all x   , f ( x)  m . Such a choice is always possible, again because of the continuity of
f ( x) . Now, consider such that x(0)   , f ( x(0))  m , and let x(t ) be the resulting trajectory.
f ( x(t ))  m . We will show that this implies that x(t )  1 . Suppose their exists t1 such that x(t1 )  1 ,
then by continuity we must have that at an earlier time t2 , x(t2 )  1 and min x 1 f ( x)  m  f ( x(t2 ))
, which is a contradiction. Thus stability in the sense of Lyapunov holds.

To prove asymptotic stability when f x(t )  c , with c  0 . We want to show that c is in fact zero.
We can argue by contradiction and suppose that c  0 . Let the set S be defined as,


S  x n

| f ( x)  c
and let Ba be a ball inside S of radius  ,

S  x  S | x   

Suppose x(t ) is a trajectory of the system that starts at x(0) , we know that f ( x(t )) is decreasing
monotonically to c and f ( x(t ))  c for all t. Therefore x(t )  Ba ; recall that Ba  S which is defined as
all the elements in n
for which f ( x)  c . In the first part of the proof, we have established that if
x(0)   then x(0)  c . We can define the largest derivative of f ( x) as,

  max f ( x )
  x c

Clearly   0 since f ( x) is lnd. Observe that,

t
f ( x(t ))  f ( x(0))   f ( x ( ))d
0

 f ( x(0))   t

Which implies that f ( x(t )) will be negative which will result in a contradiction establishing the fact
that c must be zero.
Example 1: consider the dynamical system which is governed by the differential equation.

x   g ( x)

Clearly the origin is an equilibrium point. If we define a function,


x
f ( x)   g ( y )dy
0

Then it is clear that f ( x) is locally positive definite and

f ( x)   g ( x)2

This is locally negative definite. This implies that x  0 is an asymptotically stable equilibrium point.

Lyapunov Theorem for Global Asymptotic Stability:

The region in the state space for which our earlier results hold is determined by the region over which
f(x) serves as a Lyapunov function. It is of special interest to determine the “basin of attraction" of an
asymptotically stable equilibrium point, i.e. the set of initial conditions whose subsequent trajectories
end up at this equilibrium point. An equilibrium point is globally asymptotically stable (or
asymptotically stable “in the large") if its basin of attraction is the entire state space. If a function f(x)
is positive definite on the entire state space, and has the additional property that | f  x  |  as
|| x ||  , and its derivative f is negative definite on the entire state space, then the equilibrium point
at the origin is globally asymptotically stable. We omit the proof of this result. Other versions of such
results can be stated, but are also omitted

Example 2:

Consider the nth order system

x  C ( x)

With the property that C (0)  0 and xC ( x)  0 if x  0 . Convince yourself that the unique equilibrium
point of the system is at 0. Now consider the candidate Lyapunov function

f ( x)  xx

Which satisfies all the desired properties, including f ( x)  as x  Evaluating its derivative
along trajectories, we get

A f ( x)  2 xx  2 xC ( x)  0 for x  0

Hence the system is globally asymptotically stable.

Integral curve
An Integral Curve is a parametric curve that represents a specific solution to an ordinary differential
equation or a system of equations.

dy
Let us consider a differential equation  f ( x, y ) ……………….(1)
dx

Then the graphs of the general solution of (1) are called integral curves or solution curve of equation
(1)

dy
Example: If we have,  x2  x  2
dx

Then the integral curve of the differential equation is,


Line element
The line element can be defined as a line segment associated with an infinitesimal displacement vector
in a matric space.

A short segment of the tangent line to the integral curve through (a, b) and with the slope f (a, b) is
called the line element.

In Cartesian Co-ordinates (x, y, z) the line element can be given by,

ds 2  dx 2  dy 2  dz 2

Direction field
The direction field is defined as the collection of small line segment passing through various points
having a slop that will satisfy the given differential equation at that point.

On other words, The totality of the line elements together with the corresponding directions constitute
a field that is called the direction field.
 
The equation y  f ( x, y ) gives a direction y , associated with each point ( x, y) in the plane that must
be satisfied by any solution curve passing through that point.
Null cline
dy
The null cline is the set of points in the direction field such that 0
dx

Geometrically these are the points where the vectors are horizontal algebraically. We find the null
cline by string f ( x, y)  0

Isocline
dy
An isocline is a set of points in the direction field for which there is a constant C with  C at these
dx
points. Geometrically the direction field arrows at the points of the isocline all have same slope.

Algebraically, we find the isocline for a constant C by isocline for a constant C by isocline
f ( x, y)  C

What is graphical solution and importance of graphical solution?


The graphical solution is the solution obtained from the intersecting point of the lines. The graphical
solution of the equation can be determined by using the steps given below:
Step 1: Understand the ordered pair of numbers, abscissa and ordinate in graph.
Step 2: Form the coordinates of a point by appropriate substitution.
Step 3: Locate the coordinate points obtained from Step 2 of each given equation on the graph.
Step 4: Join the plotted points by lines.
Step 5: Observe the lines of intersection, which is the solution of the equations.
The graphical solution of the equation is shown below:

The graphical view is vastly used in every type of data or report. It makes a data easier to
understand also has a lot more advantages like this. But it also has some disadvantages so for that
reason, we are giving here some advantages.

Advantages of Graphical Representation of Data

Graphical representation of reports enjoys various advantages which are as follows:

1. Acceptability: Such a report is acceptable to busy persons because it easily highlights the
theme of the report. This helps to avoid wastage of time.
2. Comparative Analysis: Information can be compared in terms of graphical representation.
Such comparative analysis helps for quick understanding and attention.
3. Less cost: Information if descriptive involves huge time to present properly. It involves more
money to print the information but the graphical presentation can be made in a short but catchy
view to make the report understandable. It obviously involves less cost.
4. Decision Making: Business executives can view the graphs at a glance and can make a
decision very quickly which is hardly possible through the descriptive reports.
5. Logical Ideas: If tables, design, and graphs are used to represent information then a logical
sequence is created to clear the idea of the audience.

Graphical Method
A procedure which gives the line element configuration of the direction field of Differential Equation
is called a graphical method.

There are two classes of graphical method

i. Direction field method


ii. Method of isoclines

(i) Direction Fields


Technique that is useful in graphing the solutions to a first-order differential equation is to sketch the
direction field for the equation. To describe this method, we need to make a general observation.
Namely a first order equation

dy
 f  x, y 
dx

Specifies a slope at each point in the xy-plane where f is defined.

A plot of short line segments drawn at various point in the xy-plane showing the slope of the solution
curve there is called a direction field for the differential equation .

(ii) Method of Isoclines


An isocline for the differential equation y '  f ( x, y) is a set of points in the xy-plane where all the
solutions have the same slope dy/dx ,thus it is a level curve for the function f  x, y 

For example, if y '  x  y , the isoclines are the straight lines x  y  C , here C is an arbitrary
constant. But C can be interpreted as the numerical value of the slope dy/dx of every solution curve as
it crosses the isocline. To implement the method of isoclines for sketching direction fields, we draw
hash marks with slope C along the isocline f ( x, y)  C for a few selected values of C. If we when
erase the underlying isocline curves, the hash marks constitute a pard of the direction field for the
differential equation.

(iii) Null clines


dx
 0. geometrically, these are the points where
The x-null cline is a set of points in the phase plane so that
dt
the vectors are either straight up or straight down. Algebraically, we find the x-null cline by solving f
(x, y) = 0.
dy
The y-null cline is a set of points in the phase plane so that  0. . Geometrically, these are the
dt
points where the vectors are horizontal, going either to the left or to the right. Algebraically, we find
the y-null cline by solving g(x, y) = 0.

Non-linear Systems
Linearization
Definition: Suppose P = (x0, y0) is an isolated critical point of the system
dx
 f ( x, y )
dt
dy
 g ( x, y )
dt

And that f and g are differentiable at P. The linearization of the system at P is the linear system
du
 f ( P)u  fx( P)v
dt
dv
 g ( P)u  fy ( P)v
dt
Written in matrix format

 du   f  x , y  f  x , y   u
 dt    x 0 0 y 0 0   
 dv   g x  x0 , y0  g y  x0 , y0    v 
 dt   

 f x  x0 , y0  f y  x0 , y0  
Where   the Jacobian matrix.
 g x  x0 , y0  g y  x0 , y0  
 

Example: The following system is non-linear.


dx
 ( x  2 y) x
dt
dy
 ( x  2) y
dt

Thinking carefully, you should be able to see that the systems has two critical points: (0, 0), (2,1) The

Jacobian is the matrix.

 0 0 
  at (0, 0)
 f x f y   2 x  2 y  2 x   0 2 
    
 gx g y   y x  2   2 4 
 1 0 
at (2,1)

The linearization and their eigenvalues at the critical points are therefore
 du
 dt  0
 at (0, 0) with 1  0, 2  2,
 dv  2v
 dt
 du
 dt  2u  4v
 at  2,1 , with 1  1  3 i, 2  1  3 i
dv
 u
 dt

Recalling our discussion of linear systems, we see that the linearization at (0,0) is degenerate, while the
linearization at (2, 1) is a spiral source (complex eigenvalues with real part positive). But what about the critical
points of the original non-linear system?


1

1

A numerically generated phase-portrait of the non-linear system


Zoomed in near(0,0) Zoomed in near (2,1)

The critical point at (2, 1) certainly looks like a spiral source, but (0, 0) just looks bizarre. This
approach of linearizing, analyzing the linearization, and piecing the results together is a standard
approach for non-linear systems. Note carefully: this is a qualitative approach; we are not solving the
non-linear system. Indeed it is probably impossible to solve this example explicitly. We should also ask
whenitislegitimatetouseinformationaboutthecriticalpointsofthelinearizationstodecidethe type of critical
point of then on-linear system. In our example, it seems legitimateat(2,1),but not so at (0, 0).

Almost Linear Systems


Definition: Anon-linear system is almost linear at an isolated critical point P=(x0, y0) if its
linearization has an isolated critical point at the origin (0,0).

Question: Identify the type of the critical point (0, 0)of the non-linear system.
 x '  2 x  3 y  4 x 2  2 xy
 '
 y  x  2 y  3xy  4 x
2

 2  8x  2 y 3  2 x   2 3 
Solution: The Jacobian is   which at  0, 0  yields   for reference, the
1  3 y  12 x 2  2x 
2
1 2 
linearization is the system

u '  2u  3v

 '
v  u  2v

The eigenvalues are   2  3i . These are both non-zero, whence (0, 0) is an isolated critical point
of the linearization. It follows that (0,0) is an isolated critical point of the original non-linear system,
andsothesystemisalmostlinearat(0,0).ApplyingtheTheorem,weseethatthecriticalpoint(0,0) is a spiral
source.

Note that the phase portraits are different, but critical points have the same type

y 1 v 1

1 1 1 1

x u

−1
−1

Almost linear system Linearization


It is not asked, but the original system can also basement be almost linear at(x,y) (0.956, 5.126) where there is a saddle
point.

Question: Find all the critical points of the non-linear system


 dx
 dt  x  y  x  xy
2


 dy   x 2  y

 dx

and identify their types. The sketch a possible phase-portrait for the system.

Solution: Tofindthecriticalpointsweneedtofindallsolutionstothesimulatanousequations
 x  y  x 2  xy  0
 2
 x  y  0

In general the reisnoguaranteed method for doing this, sober creative. Factorizing the first equation we obtain (x −
y)(1 − x) = 0, whence x = 1 or x = y. Substituting x = 1 into the second equation yields y = −1, while substituting x = y
into the second yields y2 + y = 0, whence y = 0, −1. We therefore have three critical points(1,−1),(0,0),and(−1,−1).
1  2 x  y 1  x 
The Jacobian is   . evaluating at the three critical points and computing the eigenvalues, we obtain
 2 x 1 
the following table.

Point Linearization Eigenvalues Type

 dx
 dt  u  v
 0, 0   1 Saddle
 dy  v
 dt

 du
 dt  2u
1, 1  2, 1 Nodal sink
 dv  2u  v
 dt
 du
 dt  2u  2v 1 7
(1, 1)   i Spiral source
 dv  2u  v 2 2
 dt

Since all three linearization have non-zero eigenvalues, we conclude that the original system is almost
linear at all three of its critical points. The Theorem applies and tells us that the original system has
exactly the same critical points as it linearization.

This is enough to make a very rough sketch of the


phase-portrait: the computer‟s efforts are shown.

For greater detail, we should compute the eigenvector of the


linearization sat the saddle and nodal sink in order to determine
how these are oriented. We should also identify in which
direction the spiral is rotating: in this case asking for the
direction vector at the point ( 2, 1) and observing that it is point −2 −1
in downward −suffices − oshowthatthespiralrotates counter-
clockwise. −1

−2
Theorem 2:

Proof:

Let a function f be differentiable at x  a .

f a  h  f a
Then lim exists and
h0 h

f a  h  f a 
lim  f '  a  ............................ 1
h 0 h

In order to show that f is continuous at x  a . We have to show that

lim f  x   f  a  .
x a

i.e.lim f  x   f  a   0
x a

Put x  a  h, Then as x  a, h  0 and we have show that

lim  f  a  h   f  a   0
h 0
lim f  a  h   f  a 
h 0

f a  h  f a
 lim h.... h  0, h  0
h 0 h
Now
f a  h  f a 

 lim

h 0 h
 lim

h 0

h 
 f '  a   0  0................By 1

 f is continuous at x  a .

Second part:

The converse of the statement is Even. If a function f is continuous at x  a, it need not be differentiable at x  a.
Consider the function f  x   x . It is obvious that

lim f  x   lim f  x   f  0   0
x 0 x 0

The function f is continuous at x=0.

h  0, h  h.
f 0  h  f 0 h0
For  lim  lim
h 0 h h 0 h
h
 lim  lim  1  1............................. 1
h 0 h h 0

h  0, h  h
f 0  h   f 0 h0
For  lim  lim
h 0 h h 0 h
h
 lim  lim1  1....................  2   h  0, h  0
h 0 h h 0

From (1) and (2)

f 0  h  f 0 f 0  h   f 0
lim  lim
h 0 h h 0 h
f 0  h  f 0
 lim
h 0 h

Does not exist


 f is not differentiable at x  a.

Theorem 3
Hypothesis. The roots 1 and 2 of the characteristic equation are real, unequal, and of the same sign.

Conclusion. The critical point (0,0) of the linear system is a node.

Proof. We first assume that 1 and 2 are both negative and take 1  2  0.

We consider the linear system

dx 
 ax  by 

 ................................ 1
dt
dy
 cx  dy 
dt 

The general solution of the system may then be written

x  c1 A1e1t  c2 A2e 2t


(2)
y  c1 B1e1t  c2 B2 e2t

Where A1 , B1 , A2 , and B2 are definite constants and A1 B2  A2 B1 and where c1 , c2 are arbitrary constants choosing
c2  0 we obtain the solutions

x  c1 A1e1t
(3)
y  c1 B1e1t

Choosing c1  0 we obtain the solutions

x  c2 A2e2t
(4)
y  c2 B2e2t

B1
For any c1  0 , the solutions (3) represents a path consisting of “half” of the line B1 x  A1 y of the slope . For any
A1
c1  0 , they represents a path consisting of the “other half” of this line. Since 1  0 both of these half –line paths
y B1 B
approach (0,0) as t   . Also since  these two paths enter (0,0) with slope 1 . In like manner, for any c2  0
x A1 A1
the solutions (4) represent a path consisting of “half” of the line B2 x  A2 y while for any c2  0 , the path so
represented consists of the “other half” of this line. These two half-line paths also approach (0,0) as t  

B2
And enter it with slope .
A2
Thus the solutions (3) and (4) provides us with four half-line paths which all approach and enter (0,0) as t   .

If c1  0 and c2  0 , the general solutions of (2) represents non rectilinear paths. Again since 1  2  0, all of these
paths approach (0,0) as t   .

Further, since

y c1 B1e1t  c2 B2 e2t

x c1 A1e1t  c2 A2e2t
 B1   1 2 t
 c1  e  B2
  c2 
 A1   1 2 t
 c1  e  A2
 c2 

We have

y B2
lim 
t  x A2

B2
And so all of these paths enter (0,0) with limiting slope .
A2

Thus all the paths (both rectilinear and non rectilinear ) enter (0,0) as t   and all except the two rectilinear ones
B
defined by (3) enter with slope 2 . According to the definition the critical point (0,0) is a node. Clearly it is
A2
asymptotically stable.

If now 1 and 2 are both positive and we take 1  2  0 , we see that the general solution of (1) is still of the form (2)
and particular solutions of the form (3) and (4) still exist. The situation is the same as before , except all the paths
approach and enter (0,0) as t   .

The critical point (0,0) is still a node but in this case it is clean that it is unstable.

Theorem 4

Hypothesis: The roots 1 and 2 of the characteristic equation are real, unequal and of opposite sign.

Conclusion: The critical point (0, 0) of the linear system (1) is a saddle point.

Proof: We assume that 1  0  2 . The general solution of the system (1) may again be written in the form (2) and
particular solutions of the forms (3) and (4) are again present.
For c1  0 , the solutions (3) again represent a path consisting of “half” the line B1x  A1 y ; while for any c1  0 , they
again represent a path consisting of the “other half” of the line also, since 1  0 , both of these half-line path still
approach and enter (0, 0) as t  

Figure: Qualitative diagram of the paths

Also for any c2  0 , the solutions (4) again represent a path consisting of “half” the line B2 x  A2 y ; while for any c2  0
, they again represent a path consisting of the “other half” of the line also, since 2  0 , both of these half-line path still
approach and enter (0, 0) as t  

Once again if c1  0, c2  0 , the general solution (4) represent non linear paths, but here since 1  0  2 , one of these
path approach (0, 0) as t   and t  

Thus there are two half line paths which approach and enter (0, 0) as t   and two other half –line approaches and
enter (0, 0) as t   . all other paths are no rectilinear paths which do not approach (0, 0) as t   t   , but
which become asymptotic to one or another of the four half-line paths as t   t  

Theorem 5:
Hypothesis: The roots 1 and 2 of the characteristic equation are real and equal.

Conclusion. The critical point (0,0) of the linear system is a node.

Proof: Let us first assume that 1  2    0. We consider two subcases


3(a)a  d  0, b  c  0

3(b) All other possibilities leading to a double root.

We consider first the special case 3(a). The characteristic equation becomes

 2  2a  a 2  0 and hence   a  d . The system is thus simply

dx
 x
dt
dy
 y
dt

The general solution of this system is clearly

x  c1e t
(5)
y  c2 e t

Where c1 and c2 are arbitrary constants. The paths defined by (5) for the various values of c1 and c2 are half-lines of
all possible slopes. Since   0 , we see that each of these half-lines approaches and enters (0,0) as t   . That is ,
all the paths of the system enter (0,0) as t   . According to the definition the critical point (0,0) is a node. Clearly it
is a asymptotically stable. A qualitative diagram of the paths appears in figure.

If   0 , the situation is the same except that the paths enter (0,0) as t   the node (0,0) is unstable and the arrows
in figure and all reversed.

We mention that this type of node is sometimes called a star-snapped node. Let us now consider case 3(b). Here the
characteristic equation has the double root   0 , but we exclude the special circumstances of case 3(a). The general
solution of the system (1) may in this case be written

x  c1 Aet  c2  A1t  A2  et


( 6)
y  c1 Bet  c2 ( B1t  B2 )et

B1 B
Where the A‟s and B‟s are definite constants c1 and c2 are arbitrary constants and  . Choosing c2  0 in (13.6)
A1 A
we obtain solutions

x  c1 Aet
(7)
y  c1 Bet

B
For any c1  0, the solutions (13.7) represent a path consisting of “half” of the line Bx=Ay of slope ; for any c1  0,
A
they represent a path which consists of the “other half” of this line. Since   0 both of these half-line paths approach
y B B
(0,0) as t   . Further since  both paths enter (0,0) with slope . If c2  0 the general solution (6) represents
x A A
non-rectilinear path . since   0 , we see from (13.6) that

lim x  0, lim y  0
t  t 

Thus the non rectilinear paths defined by (6) all approach (0,0) as t   . Also since

y c1 Be  c2  B1t  B2  e
t t

x c1 Aet  c2  A1t  A2  e t
 B
 c1   B2  B1t
 2
c
 A
 c1   A2  A1t
 c2 

y B1 B
We see that lim  
t  x A1 A

B
Thus all the non-rectilinear paths enter (0,0) with limiting slope .
A

B
Thus all the paths (both rectilinear and non rectilinear) enter (0,0) as t   with slope .
A

According to the definition the critical point (0,0) is a node . clearly it is a asymptotically stable.

If   0 the situation is again the same except that the paths enter (0,0) as t   ,the node (0,0) is unstable and
reversed.

Theorem 6
Hypothesis: The roots 1 and 2 of the characteristics equation are conjugate complex with real apart not zero (that is,
not pure imaginary).

Conclusion. The critical point (0, 0) of the linear system (1) is a spiral point.

Proof:

Since 1 and 2 are conjugate complex with real part not zero, we may write these roots as   i , where  and 
are both real and unequal to zero. Then by the theorem, the general solution of the system (1) may be written as,
x  e t [c1 ( A1 cos  t  A2 sin  t )  c2 ( A2 cos  t  A1 sin  t )]
( 7)
y  e t [c1 ( B1 cos  t  B2 sin  t )  c2 ( B2 cos  t  B1 sin  t )]

Where A1 A2 B1 and B2 are definite real constants and c1 and c2 are arbitrary constants.

Let us first assume that   0. Then from (1) we see that

lim x  0 and lim y  0


x  x 

and hence all paths defined by (1) approach (0, 0) as t  . We may rewrite (1) in the form

x  e t (c3 cos  t  c4 sin  t )


(8)
y  e t (c5 cos  t  c6 sin  t )

where c3  c1 A1  c2 A2 , c4  c2 A1  c1 A2 , c5  c1B1  c2 B2 , and c6  c2 B1  c1B2 . Assuming c1 and c2 are real, the solutions
(8) represent all paths in the real xy phase plane. We may now put these solutions in the form

x  K1e t cos(  t  1 )
(9)
y  K 2 e t cos(  t  2 )

Where K1  c32  c42 , K2  c52  c62 and 1 and 2 are defined by the equations

c3 c5
cos 1  , cos 2  ,
K1 K2

c4 c6
sin 1   , sin 2   .
K1 K2

Let us now consider

y K 2 e t cos(  t  2 )
 (10)
x K1e t cos(  t  1 )

K2
Letting K  and 3  1  2 , this becomes
K1

y K cos(  t  1  3 )

x cos(  t  1 )
 cos(  t  1 ) cos 3  sin(  t  1 ) sin 3 
K  (11)
 cos(  t  1 ) 
 K [cos 3  sin 3 tan(  t  1 )]
Provided cos(  t  1 )  0. As a result of the periodicity of the trigonometric functions involved in ($) and (11) we
y
conclude from these expression that lim does not exist and so the paths do not enter (0, 0). Instead, it follows from
x  x

(10) and (11) that the paths approach (0, 0) in a spiral-like manner, winding around (0, 0) an infinite number of times
as t  . According to the definition, the critical point (0, 0) is a spiral point. Cleary, it is asymptotically stable. A
Qualitative diagram of the paths appears in figure.

If   0, the situation is the same except that the paths approach (0, 0) as t   the spiral point (0, 0) is unstable,
and the arrows in figure are reversed.

Theorem 7

Hypothesis: The roots 1 and 2 of the characteristic equation are pure imaginary.

Conclusion: The critical point (0, 0) of the linear system is a center.

Proof: Since 1 and 2 are pure imaginary we may write them as    i . Where   0 and  is real and unequal to
zero .then the general solution of the system is for, where   0 . In the notation of all real solution can be written as the
form,

x  K1 cos(  t  1 )
(12)
y  K 2 cos(  t  2 )

Where, K1, K2 ,1,2 are defined as before. The solutions (12) define the paths in the real xy phase plane. Since the
trigonometric functions in (12) oscillate identically between 1ast   and as t  
Figure: Qualitative diagram of the paths

Hence the path are the closed curves surroundings (0, 0), members of which are arbitrarily close to (0, 0). Indeed they
are in an infinite family of ellipses with centre at (0, 0). According to definition of this section. The critical point is a
centre. Clearly it is stable. However, since the path do not approach (0, 0), the critical point is not asymptotically stable.
A qualitative diagram of the paths appears in the figure.

Theorem 8
Consider the linear system

dx
 ax  by
dt (13)
dy
 cx  dy
dt

Where ad  bc  0 so that (0, 0) is the only critical point of the system.

1. If both roots of the characteristic equation are real and negative or conjugate complex with negative real
parts, then the critical point (0, 0) of (13.19) is asymptotically stable.
2. If the roots of are pure imaginary, then the critical point (0, 0) of (1) is stable, but not asymptotically stable
3. If either of the root is real and positive or if the roots are conjugate complex with positive real parts, then the
critical point (0, 0) of (1) is unstable.

Nature of roots of 1 and 2 Nature of critical point (0, 0) Stability of critical point (0, 0)
characteristic equation of linear system
 2  (a  d )  (ad  bc)  0 dx
 ax  by
dt
dy
 cx  dy
dt
Real, unequal and same sign Node Asymptotically stable if roots
are negative; unstable if roots
are positive
Real, unequal and opposite Saddle point Unstable
sign

Real and equal Node Asymptotically stable if roots


are negative; unstable if roots
are positive
Conjugate complex but not Spiral point Asymptotically stable if roots
pure imaginary are negative; unstable if real
part of roots are positive

Pure imaginary Center Stable, but not asymptotically


stable.

Example of Null clines

Consider the system

dx  x
 2 x 1    xy
dt  2
dy  y
 3 y 1    2 xy
dt  3
 x
To find the -nullcline, we solve 2 x 1    xy  0 , where multiplying out and collecting the common factor of
 2
gives x  2  x  y   0 . This gives two -nullclines, the line x  y  2 and the

𝑥
1 2 3

 x
-axis. By plugging in the points 1, 0  and  2, 2  into 2 x 1    xy , we see that solutions (to the left of the -axis)
 2
move to the right if below the line x  y  2 and to the left if above it.
 y
To find -nullcline, we solve 3 y 1    2 xy  0 , where multiplying out and collecting the common factor of gives
 3

y  3  y  2 x   0 . This gives two -nullclines, the line 2 x  y  3 and the -axis. By plugging in the points  0,1 and

 y
 2, 2  into 3 y 1    2 xy , we see that solutions (to the left of the -axis) move to the right if below the line
 3
2 x  y  3 and down if above it.
𝑦

𝑥
1 2 3

Combining this information gives us the following picture. Notice that we can draw directions on each nullcline by
using the direction information from the other graph. For example, the line segment from 1,1 to  0,3  , since it is

above the line x  y  2 , has solutions moving to the left.


𝑦

𝑥
1 2 3

dx dy
Also, where the -nullcline and -nullcline cross, both and are zero. So these points (marked by dots in the
dt dt
above graph) are equilibrium points.

Once a solution enters the triangle with vertices 1,1 ,  0, 2 and  0,3  , it can never leave. Similarly, solutions in the

3 
triangle with vertices 1,1 ,  , 0  .
2 

Problem 1: Find the equilibrium points and discuss the null-clines of the system
 
x  x( x  3 y  150) , y  y( 2 x  y  100)

Solution: We have the given system as,


x  x( x  3 y  150) ---------------------------------(i)

y  y (2 x  y  100) --------------------------------(ii)

dx
The x-null clines are given by, 0
dt

So from (i), x( x  3 y  150)  0


Then x  0 .............. (iii) and  x  3 y  150  0
 x  3 y  150 .............. (iv)
dy
Again the y-null clines are given by, 0
dt

So from (ii), y(2 x  y  100)  0


Then, y  0 ................ (v) and 2 x  y  100  0
 2 x  y  100 ................ (vi)

Hence from (iii), (iv), (v) and (vi) the equilibrium points are,

(0,0);(0,50);(150,0);(50,0);(0,100);(30, 40)

Problem 2: Find the equilibrium points and discuss the null-clines of the system
 
x  x(10  x  y ) , y  y (30  2 x  y)

Solution: We have the given system as,



x  x(10  x  y ) ------------------------------------(i)

y  y (30  2 x  y ) ---------------------------------(ii)

dx
The x-nullclines are given by, 0
dt

So from (i), x(10  x  y)  0


Then x  0 ...................... (iii) and 10  x  y  0
 x  y  10 .................. (iv)

dy
Again the y-nullclines are given by, 0
dt

So from (ii), y(30  2 x  y)  0


Then, y  0 ................... (v) and 30  2 x  y  0
 2 x  y  30 ...................... (vi)

Hence from (iii), (iv), (v) and (vi) the equilibrium points are,

(0,0);(10,0);(0,10);(15,0);(0,30);(20, 10)

Problem 3: Find the equilibrium points and discuss the null-clines of the system
 x  9
x  2 x(1  )  xy , y  y (  y 2 )  x 2 y
2 4

Solution: We have the given system as,


 x
x  2 x(1  )  xy ----------------------------------(i)
2
 9
y  y (  y 2 )  x 2 y -------------------------------(ii)
4
dx
The x-nullclines are given by, 0
dt

x
So from (i), 2 x(1  )  xy  0
2
x
 x{2(1  )  y}  0
2
 x(2  x  y )  0
Then x  0 ................. (iii) and 2  x  y  0
 x  y  2 ................... (iv)
dy
Again the y-nullclines are given by, 0
dt

9
So from (ii), y (  y 2 )  x 2 y  0
4
9
 y(  y 2  x2 )  0
4
9
Then, y  0 ................... (v) and  y 2  x 2  0
4
2
3
 x  y    ................... (vi)
2 2

2
Hence from (iii), (iv), (v) and (vi) the equilibrium points are,

3 3  4 2 4 2 
(0, 0);(2, 0);(0, 2);(  , 0);(0,  );  , 
2 2  4 4 

Problem 4: Find the equilibrium points and discuss the null-clines of the system
 
x  x(4 x  y  160) , y  y ( x 2  y 2  2500)

Solution: We have the given system as,



x  x(4 x  y  160) -----------------------------(i)

y  y ( x 2  y 2  2500) -------------------------(ii)
dx
The x-nullclines are given by, 0
dt

So from (i), x(4 x  y  160)  0

Then x  0 ................... (iii) and 4 x  y  160  0


 4 x  y  160 .................... (iv)

dy
Again the y-nullclines are given by, 0
dt

So from (ii), y ( x 2  y 2  2500)  0

Then, y  0 .................... (v) and  x 2  y 2  2500  0


 x 2  y 2   50  ...................... (vi)
2

Hence from (iii), (iv), (v) and (vi) the equilibrium points are,

(0, 0);(40, 0);(0,160);(50, 0);(0, 50); 30, 40  ;(45.294, 21.176)

Problem 5: Find the equilibrium points and discuss the null-clines of the system
dx  x
 8 x 1    xy
dt  8

dy
 4 y  2 xy
dt

Solution: We have the given system as,

dx  x
 8 x 1    xy …………………….....(i)
dt  8

dy
 4 y  2 xy ………………………………(ii)
dt

dx
The x nullclines are given by, 0
dt

So from equation (i),

 x
8 x 1    xy  0
 8

 x(8  x  y)  0

 x  0 …………….(iii) or, 8  x  y  0

 x  y  8  0 ……………..…(iv)

dy
Again y nullclines are given by, 0
dt

So from equation (ii)

4 y  2 xy  0

 y(4  2 x)  0

 y  0 ………………(v) Or, 4  2 x  0 ………………..…(vi)

Hence from (iii), (iv), (v) and (vi) the equilibrium points are,

(0,0), (8,0), (2,6)


Problem 6: Find the equilibrium points and discuss the null-clines of the system

dx  x
 2 x 1    xy
dt  2

dy  y
 y 1    3xy
dt  4

Solution: We have the given system as,

dx  x
 2 x 1    xy …...........................(i)
dt  2

dy  y
 y 1    3xy ……………..…….(ii)
dt  4

dx
The x nullclines are given by, 0
dt

So from equation (i)

 x
2 x 1    xy  0
 2

 x(2  x  y)  0

 x  0 ………………(iii) or, 2  x  y  0

 x  y  2  0 …………………(iv)
dy
Again y nullclines are given by, 0
dt

So from equation (ii),

 y
y 1    3xy  0
 4

 y(4  y  3x)  0

 y  0 …………………(v) Or, 4  y  3x  0

 3x  y  4  0 ...........................(vi)

Now from (3),(4),(5) and (6) we get the equilibrium points are, (0,0), (0,4), (2,0), (1,1)

Problem 7: Find the equilibrium points and discuss the null-clines of the system

dx
 x(1  x)  xy
dt
dy y
 2 y (1  )  3xy
dt 2

Solution: We have the given system as,


dx
 x(1  x)  xy.................(i )
dt
dy y
 2 y (1  )  3xy...........(ii )
dt 2

dx
The x-nullclines are given by, 0
dt

So from (i),

x(1  x)  xy  0
 x(1  x  y )  0

(1  x  y )  0
So, x = 0 ……………………….…(iii) or,
 x  y  1  0.......................(iv )

dy
Again the y-nullclines are given by, 0
dt

So from (ii),

y
2 y (1  )  3xy  0
2
 y (2  y  3x)  0

2  y  3x  0
So, y  0.....................(v) or,
 3x  y  2  0..................(vi )

1 1
Hence from (iii), (iv), (v) and (vi) we get the equilibrium points are, (0, 0), (0, 2), (1, 0),  ,  .
2 2
Problem 8: Find the equilibrium points and discuss the null-clines of the system

dx
 x(1  x)  xy
dt
dy y2
 2 y (1  )  3x 2 y
dt 2

Solution: We have the given system as,

dx
 x(1  x)  xy....................(i )
dt
dy y2
 2 y (1  )  3 x 2 y...........(ii )
dt 2

dx
The x-nullclines are given by, 0
dt

x(1  x)  xy  0
So from (i),
 x(1  x  y )  0

(1  x  y )  0
So, x = 0 ………………………(iii) or,
 x  y  1  0.....................(iv)
dy
Again the y-nullclines are given by, 0
dt

y2
2 y (1  )  3 x 2 y  0
So from (ii), 2
 y (2  y 2  3x 2 )  0

2  y 2  3x 2  0
So, y  0.....................(v) or,
 3x 2  y 2  2  0...................(vi)

Hence from (iii), (iv), (v) and (vi) we get the equilibrium points are,
 1 5 3  5   1 5 3  5 
 
(0, 0), (1, 0), 0, 2 ,  , ,
4   4
, .
4 
 4

Method of Isoclines
An isocline for the differential equation y′ = f (x, y) is a set of points in the xy-plane where all the solutions have the
same slope dy/dx
Method of isoclines for sketching direction fields STEPS:
1. we draw different isoclines with different c
2. As c is the numerical value of the slope of every solution curve as it crosses the isocline, we draw hash
marks with slope c along the isocline ( ) for a few selected values. The hash marks constitute a part of the
direction field for the differential equation.

Example of Isocline
Example 1: ,

The isoclines are the straight lines


Mark slope c on each isocline (there are some between isoclines here too)

,
Example 3: Draw the graph of the following two isoclines in MATLAB .

dx Lx
 f  x, y   A  Bx 
dt 1  Mx 1  Ny 
dy Lx F
 g  x, y   D
dt 1  Mx 1  Ny  1  Hy 
Where A  2; B  0.05; L  0.004; M  0.004; N  0.002; D  0.053; F  0.02; H  0.0004

Solution: A combination of „meshgrid‟ and „surf‟ functions maybe used in order to do this. The following code snippet
may or may not be accurate, but feel free to use it as a starting point.
A=2;B=0.05;L=0.004;M=0.004;N=0.002;D=0.053;F=0.02;H=0.0004
f=@(x,y)A-B.*x-L.*x./(1+M.*x).*(1+N.*y)
g=@(x,y)L.*x./(1+M.*x).*(1+N.*y)-D-F./(1+H.*y)
[X,Y]=meshgrid(1:5,1:5);
surf(X,Y,f(X,Y));
hold on;
surf(X,Y,g(X,Y));
Example 4: Draw the graph of the following two isoclines in MATLAB .

dx
 f  x, y   x  A  Bx  Ly 
dt
dy
 g  x, y   y   M  Nx  Dy 
dt

Where A  2; B  1; L  1; M  1; N  1; D  1

Solution: A combination of „meshgrid‟ and „surf‟ functions maybe used in order to do this. The following code snippet
may or may not be accurate, but feel free to use it as a starting point.
A=2;B=1;L=1;M=1;N=1;D=1
f=@(x,y)x.*(A-B.*x-L.*y)
g=@(x,y)y.*(-M+N.*x-D.*y)
[X,Y]=meshgrid(1:5,1:5);
surf(X,Y,f(X,Y));
hold on;
surf(X,Y,g(X,Y));
dy 3 x  y
Example 5: Use the method of isoclines to sketch several integral curves of 
dx x  y

dy 3 x  y
Solution: We have  …………. (1)
dx x  y

3x  y
The isocline of (1) are given by  c …………………(2)
x y

Where c represents slope of the integral curves of (1) . The equation (2) represents a family of straight lines. We draw
the curves of (2) for c  0, 1, 2,...

Thus taking c  tan  , Where  is the inclination of the line element.

When c  0 then y  3x   tan 1 c  00

When c  1 then y   x   tan 1 c  450


When c  1 then x  0   tan 1 c  450

1
When c  2 then y   x   tan 1 c  630
3

When c  2 then y  5 x   tan 1 c  630

When c   then y  x   tan 1 c  900

dy
Example 6: Use the method of isoclines to sketch several integral curves of  x2  2 y2
dx

dy
Solution: We have  x 2  2 y 2 …………. (1)
dx

dy
The isocline of (1) are given by  x 2  2 y 2 …………………(2)
dx

Where c represents slope of the integral curves of (1) . The equation (2) represents a family of straight lines. We draw
the curves of (2) for c  1, 2,3, and 4

Thus taking c  tan  , Where  is the inclination of the line element.

When c  1 then x  2 y  1   tan 1 c  450


2 2

When c  2 then x  2 y  2   tan 1 c  630


2 2

When c  3 then x  2 y  3   tan 1 c  720


2 2

When c  4 then x  2 y  1   tan 1 c  750


2 2

dy
Example 7: Use the method of isoclines to sketch several integral curves of  3x  y
dx

dy
Solution : We have,  3 x  y ……………………………………………………(1)
dx

The isoclines of (1) are given by 3x  y  c ……………………………(2)

Where c represents the slope of integral curves of (1). The equation (2) represents a family of straight lines. We draw
the curves of (2) for c  0, 1, 2
Thus taking c  tan  where  is the inclination of line element.

When c= 0 then y=3x  =tan 1c  0


When c= 1 then y= 3x-1  =tan 1c  450
When c= -1 then y=3x+1  =tan 1c  450
When c= 2 then y=3x-2  =tan 1c  630
When c= -2 then y=3x+2  =tan 1c  630

On each of these curves we construct a number of the elements with approximate inclination.

dy y  x
Example 8: Use the method of isoclines to sketch several integral curves of 
dx x  y

dy y  x
Solution : We have,  ……………………………(1)
dx x  y

yx
The isoclines of (1) are given by  c ………………………(2)
x y

Where c represents the slope of integral curves of (1). The equation (2) represents a family of straight lines. We draw
the curves of (2) for c  0, 1, 2

Thus taking c  tan  where  is the inclination of line element.

When c= 0 then y= x  =tan 1c  0


When c= 1 then x= 0  =tan 1c  450
When c= -1 then y= 0  =tan 1c  450
When c= 2 then y=-3x  =tan 1c  630
1
When c= -2 then y=- x  =tan 1c  630
3
When c=  then y= -x  =tan 1c  900

On each of these curves we construct a number of the elements with approximate inclination.

Qualitative Analysis
Very often it is almost impossible to find explicitly of implicitly the solutions of a system (specially nonlinear ones). The
qualitative approach as well as numerical one are important since they allow us to make conclusions regardless whether
we know or not the solutions.

Example 1: Consider the model describing two species competing for the same prey
dx
 x (1  x )  xy
dt
dy
 2 y(1  y / 2)  3xy
dt

Let us only focus on the first quadrant and . First, we look for the equilibrium points. We must have
x (1  x )  xy  0
2 y(1  y / 2)  3xy  0

Algebraic manipulations imply


x0 or 1  x  y  0

and
y0 or 2  3x  y  0

The equilibrium points are (0,0),(0,2),(1,0),and .

Consider the region R delimited by the x-axis, the y-axis, the line 1-x-y=0, and the line 2-3x-y=0.
Hence the direction of the motion is the same (that is left-down) as long as the trajectory lives inside this region.

Infect, looking at the first-quadrant, we have three more region stood to the above one. The direction of the
motion depends on what region we are in (see the picture below)

The boundaries of the seregionsareveryimportantindeterminingthedirectionofthemotionalongthetrajectories.Infact,


ithelpsto visualizethetrajectoriesasslope-fielddidforautonomousequations.Theseboundariesarecallednull clines.

Null clines.

Consider the autonomous system

dx
 f ( x , y)
dt
dy
 g ( x , y)
dt
Between x-null cline and y-null cline are exactly the equilibrium points. Note that along the x-null cline the
velocity vectors are vertical while along the y-null
clinethevelocityvectorsarehorizontal.Notethataslongaswearetravelingalonganull cline without crossing an
equilibrium point, then the direction of the velocity vector must be the same. Once we cross an equilibrium point,
then we may have a change in the direction(from up to down, or right to left, and vice-versa).

Example 2 : Draw the null clines, for the autonomous system and the velocity vectors along them.
dx
 x (1  x )  xy
dt
dy
 2 y(1  y / 2)  3xy
dt

The x-null cline are given by


dx
 x (1  x )  xy  0
dt

which is equivalent to

x  0 or 1  x  y  0

while the y- null cline, are given by

dy
 2 y(1  y / 2)  3xy  0
dt
which is equivalent to

y  0 or 2  3x  y  0

In order to find the direction of the velocity vectors along the null clines, we pick a point on the
null clines, and find the direction of the velocity vector at that point. The velocity vector along the
segment of the null clines,delimitedbyequilibriumpointswhichcontainsthegivenpointwill
havethesamedirection.Forexample,considerthepoint(2,0).Thevelocityvectoratthispointis(-

1,0).Thereforethevelocityvectoratany point(x,0), with x>1,ishorizontal(wearer on they)null clines,


and points to the left. The picture below gives the null clines, and the velocity vectors along them.

In this example, the null clines, are lines. In general we may have any kind of curves.
Example 3: Draw the null clines for the autonomous system

dx
 x (1  x )  xy
dt
dy
 2 y(1  y 2 / 2)  3x 2 y
dt
The x-null cline are given by
dx
 x (1  x )  xy  0
dt

which is equivalent to

x  0 or 1  x  y  0

dx
 2 y(1  y 2 / 2)  3x 2 y
dt

Which is equivalent to
y  0 or 2  3x 2  y 2  0

Hence the y-null cline is the union of a line with the ellipse
3x 2  y 2  0

Example 4: Discuss the behavior of the solutions of the autonomous system


dx
 x (1  x )  xy
dt
dy
 2 y(1  y / 2)  3xy
dt
We have already found the null clines and the direction of the velocity vectors along these null
clines.

These null clines, give the birth to four region sin which the direction of the motion is constant. Let
us discuss the region border end by the x- axis,they-axis,theline1-x-y=0,andtheline2-3x-
y=0.Thenthedirectionofthemotionisleft-down.Soamovingobjectstartingata
positioninthisregion,willfollowapathgoingleft-down.Wehavethreechoices

First choice: the trajectory dies at the equilibrium point  1 , 1  .


 3 3

Second choice: the starting point is above the trajectory which dies at the equilibrium point

Then the trajectory will hit the

Triangle defined by the points , ,(0,1),and(0,2).Then it will group-left and dies at the
equilibrium point(0,2).

Third choice: the starting point is below the trajectory which dies at the equilibrium point. Then the
trajectory will hit the

Triangle defined by the points , , (1,0),and .Then it will go down-right and dies at the
equilibrium point(1,0).
For the other regions, look at the picture below. We included some solutions for every region.

We see from this example that the trajectories which dye at the equilibrium point are crucial to
predicting the behavior of

The solutions . The set we trajectories are called separatrix because they separate the region sin to
different sub regions with a specific behavior. To find them is a very difficult problem. Notice also that
the equilibrium points (0,2) and (1,0) behave like sinks. The classification of equilibrium points will be
discuss reducing the approximation by linear system.

Example 5: Consider the first order system

1.
Graph the nullclines for this system.
2.
Insert direction field "arrows" at the nullclines
3.
Determine the general direction of the direction field in each of the regions cut out by the nullclines.
4.
Sketch possible phase portraits for the two initial conditions indicated by "dots" for as long as the
phase portraits stay within the graph's frame.

5.
Find all the equilibrium points of the system.
Answer:
1.
The x-nullclines can be obtained by setting x'=0, i.e.

The y-nullclines can be obtained by setting y'=0, i.e.


For 2, 3, 4, use the following graph:

5.
The equilibrium points of the system are the intersection points of the x-nullclines and y-nullclines.
Therefore, we get
,

which yields x=2 and . Hence, we have two equilibrium points


.

Example 6: Consider the system

.
Draw the nullclines and find all the equilibrium points. Use this information to determine the fate of
the solutions corresponding to the following initial conditions:

Answer: Note that this system is non linear. The x-nullclines are given by

We recognize the two lines x=0 (y-axis) and the line -x-y+1=0.
The y-nullclines are given by
We recognize the line y=0 (x-axis) and the circle (centered at (0,0) with radius
2). In the graph below we give the nullclines:

The equilibrium points may be obtained as the intersection between the x-nullclines and y-nullclines.
There are 6 points (see the above graph). The solutions corresponding to the given points are drawn
in white in the graph below:
Example 7: Consider the system

Find the equilibrium points and the nullclines. Draw the vector field. Sketch some solutions and
specially the solutions around the equilibrium points.
Answer. The equilibrium points are given by the algebraic system

It is easy to see that we must have x=0 (from the second equation) and therefore y=0. Hence the
system has one equilibrium point (0,0). The x-nullcline are given by the equation

which is the graph of the function . The y-nullcline are given by the equation
which reduces to the line x=0 (the y-axis). In the picture below, we draw the vector field as well as
the nullclines.

Clearly, the solutions spiral around the equilibrium point (see the picture below)

Notice that the solutions spiral and get closer to a cycle. We can see this better by looking at the
graphs of x versus t as well as the graphs of y versus t.
and

Example 8: Consider the system

Find the equilibrium points and the nullclines. Draw the vector field. Sketch some solutions and
specially the solutions around the equilibrium points.
Answer. The equilibrium points are given by the algebraic system

From the first equation, we get y=0. The second equation gives x=0 or x=1. Hence the equilibrium
points are (0,0) and (1,0). The x-null cline are given by the equation y = 0 which is the x-axis and the y-

null cline are given by the equation , which reduces to the two vertical lines x=0 (the y-
axis) and x=1. In the picture below, we draw the vector field as well as the nullclines.
Clearly, the solutions spiral around the equilibrium point (1,0) and get away from the other
equilibrium point (0,0)(see the picture below)

A closer look at the solutions around the equilibrium point (1,0) gives

Clearly the solutions are close to cycles. The graphs of x versus t as well as the graphs of y versus t
illustrate better this remark:
and

You might also like