M
M[3 || 4]A21 Mathematical Biology: Example sheet 1 A.J. Mestel
Population Dynamics of a single species; Fisher's equation.
1. An organism, with population density U(X, T), has a logistic growth and diffuses in space, so that
UT = R U(1- |
U Q
|
)+DUXX Î a < X < b , |
|
where R, Q and D are known positive constants.
Defining x, t and u(x, t) appropriately, obtain a non-dimensional version of this equation
ut = u(1-u)+uxx Î 0 < x < L , (1) |
|
and give a formula for L.
2. The population density u(x,t) obeys the Fisher equation (1) together with the
boundary conditions
,
u(0, t) = 0 = u(L, t) u(x, 0) = u0(x) ³ 0 . |
|
You may assume that u(x, t) ³ 0 for all x and t.
Defining a sort of ``energy," E(t) = 1/2ò0L u2 dx, show that
|
dE dt
|
= |
ó õ
|
L
0
|
u2(1-u) dx- |
ó õ
|
L
0
|
ux2 dx . |
|
Assuming the Poincaré inequality for smooth functions f(x) satisfying f(0) = f(L) = 0,
|
ó õ
|
L
0
|
fx2 dx ³ |
p2 L2
|
|
ó õ
|
L
0
|
f2 dx , (Poincaré) |
|
deduce that the population will die out if L < p. Write down the corresponding
result in terms of the dimensional quantities a, b, R, D and Q
of question 1.
3. Suppose, instead, that u(x,t) is subject to the `no flux' boundary conditions
,
ux(0, t) = 0 = ux(L, t) u(x,0) = u0(x) . |
|
This time consider F(t) = 1/2ò0L ux2 dx. Show that if L < p, then F®0 as t®¥, and hence u must tend to a constant for any initial state u0(x).
[Hint: First differentiate (1) with respect to x, and continue as in the last question.]
4. A swimming pool occupies 0 < x < a, 0 < y < b, 0 < z < c. The density of algae
within it, u(x, y, z, t), is assumed to obey the equation and boundary conditions
ut = u(1-u)+Ñ2u with u = 0 on x = 0,a, y = 0,b, z = 0,c . |
|
Discuss whether or not the `clean' state u º 0 is stable.
Is this a good model, do you think?
File translated from TEX by TTH, version 2.22.
On 14 Oct 1999, 16:07.