Differential Equations

Copyright © Cengage Learning. All rights reserved.

9.4 Models for Population Growth

Copyright © Cengage Learning. All rights reserved.

Models for Population Growth
In this section we investigate differential equations that are
used to model population growth: the law of natural growth,
the logistic equation, and several others.

3
The Law of Natural Growth

4
The Law of Natural Growth
In general, if P (t) is the value of a quantity y at time t and if
the rate of change of P with respect to t is proportional to its
size P (t) at any time, then

where k is a constant.

Equation 1 is sometimes called the law of natural growth.

If k is positive, then the population increases; if k is
negative, it decreases.
5
The Law of Natural Growth
Because Equation 1 is a separable differential equation, we
can solve it by the methods given below:

ln | P | = kt + C

| P | = ekt + C = eCekt

P = Aekt

where A (= eC or 0) is an arbitrary constant.

6
The Law of Natural Growth
To see the significance of the constant A, we observe that

P (0) = Aek  0 = A

Therefore A is the initial value of the function.

7
The Law of Natural Growth
Another way of writing Equation 1 is

which says that the relative growth rate (the growth rate
divided by the population size) is constant.

Then says that a population with constant relative

growth rate must grow exponentially.

8
The Law of Natural Growth
We can account for emigration (or “harvesting”) from a
population by modifying Equation 1: If the rate of emigration
is a constant m, then the rate of change of the population is
modeled by the differential equation

9
The Logistic Model

10
The Logistic Model
As we studied earlier, a population often increases
exponentially in its early stages but levels off eventually
and approaches its carrying capacity because of limited
resources.

If P(t) is the size of the population at time t, we assume that

if P is small

This says that the growth rate is initially close to being

proportional to size.

11
The Logistic Model
In other words, the relative growth rate is almost constant
when the population is small. But we also want to reflect
the fact that the relative growth rate decreases as the
population P increases and becomes negative if P ever
exceeds its carrying capacity M, the maximum population
that the environment is capable of sustaining in the long
run.

The simplest expression for the relative growth rate that

incorporates these assumptions is

12
The Logistic Model
Multiplying by P, we obtain the model for population growth
known as the logistic differential equation:

13
Example 1
Draw a direction field for the logistic equation with k = 0.08
and carrying capacity M = 1000. What can you deduce
about the solutions?

Solution:
In this case the logistic differential equation is

14
Example 1 – Solution cont’d

A direction field for this equation is shown in Figure 1.

Direction field for the logistic equation in Example 1

Figure 1

15
Example 1 – Solution cont’d

We show only the first quadrant because negative

populations aren’t meaningful and we are interested only in
what happens after t = 0.

The logistic equation is autonomous (dP/dt depends only

on P, not on t), so the slopes are the same along any
horizontal line. As expected, the slopes are positive for
0 < P < 100 and negative for P > 1000.

The slopes are small when P is close to 0 or 1000 (the

carrying capacity). Notice that the solutions move away
from the equilibrium solution P = 0 and move toward the
equilibrium solution P = 1000.
16
Example 1 – Solution cont’d

In Figure 2 we use the direction field to sketch solution

curves with initial populations P (0) = 100, P (0) = 400, and
P (0) = 1300.

Solution curves for the logistic equation in Example 1

Figure 2

17
Example 1 – Solution cont’d

Notice that solution curves that start below P = 1000 are

increasing and those that start above P = 1000 are
decreasing.

The slopes are greatest when P  500 and therefore the

solution curves that start below P = 1000 have inflection
points when P  500.

In fact we can prove that all solution curves that start below
P = 500 have an inflection point when P is exactly 500.

18
The Logistic Model
The logistic equation is separable and so we can solve it
explicitly. Since

we have

19
The Logistic Model
To evaluate the integral on the left side, we write

Using partial fractions, we get

This enables us to rewrite Equation 5:

20
The Logistic Model

where A = e–C.

21
The Logistic Model
Solving Equation 6 for P, we get

so

22
The Logistic Model
We find the value of A by putting t = 0 in Equation 6. If t =
0, then P = P0 (the initial population), so

Thus the solution to the logistic equation is

23
The Logistic Model
Using the expression for P(t) in Equation 7, we see that

which is to be expected.

24
Example 2
Write the solution of the initial-value problem

P (0) = 100

and use it to find the population sizes P(40) and P(80). At

what time does the population reach 900?

25
Example 2 – Solution
The differential equation is a logistic equation with k = 0.08,
carrying capacity M = 1000, and initial population P0 = 100.
So Equation 7 gives the population at time t as

where

Thus

So the population sizes when t = 40 and 80 are

26
Example 2 – Solution cont’d

The population reaches 900 when

27
Example 2 – Solution cont’d

Solving this equation for t, we get

So the population reaches 900 when t is approximately 55. 28

Example 2 – Solution cont’d

As a check on our work, we graph the population curve in

Figure 3 and observe where it intersects the line P = 900.

Figure 3

The cursor indicates that t  55.

29
Comparison of the Natural Growth
and Logistic Models

30
Comparison of the Natural Growth and Logistic Models

In the 1930s the biologist G. F. Gause conducted an

experiment with the protozoan Paramecium and used a
logistic equation to model his data. The table gives his daily
count of the population of protozoa.

He estimated the initial relative growth rate to be 0.7944

and the carrying capacity to be 64.

31
Example 3
Find the exponential and logistic models for Gause’s data.
Compare the predicted values with the observed values
and comment on the fit.

Solution:
Given the relative growth rate k = 0.7944 and the initial
population P0 = 2, the exponential model is

P (t) = P0ekt

= 2e0.7944t

32
Example 3 – Solution cont’d

Gause used the same value of k for his logistic model. [This
is reasonable because P0 = 2 is small compared with the
carrying capacity (M = 64).

The equation

shows that the value of k for the logistic model is very close
to the value for the exponential model.]

33
Example 3 – Solution cont’d

Then the solution of the logistic equation in Equation 7

gives

where

So

34
Example 3 – Solution cont’d

We use these equations to calculate the predicted values

(rounded to the nearest integer) and compare them in the
following table.

35
Example 3 – Solution cont’d

We notice from the table and from the graph in Figure 4

that for the first three or four days the exponential model
gives results comparable to those of the more sophisticated
logistic model.

For t  5, however, the

exponential model is
hopelessly inaccurate, but
the logistic model fits the
observations reasonably
well. The exponential and logistic
models for the Paramecium data
Figure 4

36
Other Models for
Population Growth

37
Other Models for Population Growth
The Law of Natural Growth and the logistic differential
equation are not the only equations that have been
proposed to model population growth.

Two of the other models are modifications of the logistic

model. The differential equation

has been used to model populations that are subject to

harvesting of one sort or another. (Think of a population of
fish being caught at a constant rate.)
38
Other Models for Population Growth
For some species there is a minimum population level m
below which the species tends to become extinct. (Adults
may not be able to find suitable mates.) Such populations
have been modeled by the differential equation

where the extra factor, 1 – m/p, takes into account the

consequences of a sparse population.

39