Math 5110/6830
Instructor: Alla Borisyuk
Homework 3.2
Due: September 15
The discrete-time logistic model that we have considered in class is from a
family of models that can be written as
n+1 = g(xn )xn :
x
If n is the population size, then ( n ) can be interpreted as the growth rate
(number of osprings per adult). If ( n ) 6= const, then the growth rate is
dependent on the current population size or density-dependent. Dierent models
will assume dierent type of dependence (form of ( )). Logistic map is one
possible model with linear form of ( ). It has proven useful and is considered
classical. However, as we discussed in class, not all of its solutions exhibit logistic
growth (that we saw in the data) and for some parameter values the model does
not make sense at all. In particular, for n
we have ( n ) 0, which
should not happen. The models below overcome this by choosing dierent form
of ( n ).
1. The Beverton-Holt Model. Consider
x
g x
g x
g x
g x
x
> K
g x
<
g x
( n) =
1 + rK1 n
r
g x
;
x
with
0 and
0.
a)Plot the graph of ( n ) for some values of
and and notice the dierence
with the logistic model.
b)Find xed points of this map.
c)Study analytically the stability of the xed points.
c)Verify your results from b) with cobwebbing (you'll need to draw separate
pictures for several regions of ).
d) Use your cobweb diagrams to sketch the solutions
r >
K >
g x
K
r
r
2.
The Ricker Model.
For this model
( n) =
g x
h
exp
r
(1
n )i ;
x
K
with
0 and
0.
a)Plot the graph of ( n ) and notice the dierence with the logistic model and
similarity and dierence with the Beverton-Holt model. b)Find xed points of
this map.
c)Study analytically the stability of the xed points.
d)Do cobwebbing and sketch the solution for some 0
2.
r >
K >
g x
< r <
1