forked from johannesgerer/jburkardt-f
-
Notifications
You must be signed in to change notification settings - Fork 1
/
fd_predator_prey.html
326 lines (280 loc) · 9.61 KB
/
fd_predator_prey.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
<html>
<head>
<title>
FD_PREDATOR_PREY - Finite Difference Solution of a Predator Prey ODE System
</title>
</head>
<body bgcolor="#EEEEEE" link="#CC0000" alink="#FF3300" vlink="#000055">
<h1 align = "center">
FD_PREDATOR_PREY <br> Finite Difference Solution of a Predator Prey ODE System
</h1>
<hr>
<p>
<b>FD_PREDATOR_PREY</b>
is a FORTRAN90 program which
applies the finite difference method to estimate solutions of a
pair of ordinary differential equations that model the behavior of
a pair of predator and prey populations.
</p>
<p>
The physical system under consideration is a pair of animal populations.
</p>
<p>
The PREY reproduce rapidly; for each animal alive at the beginning of the
year, two more will be born by the end of the year. The prey do not have
a natural death rate; instead, they only die by being eaten by the predator.
Every prey animal has 1 chance in 1000 of being eaten in a given year by
a given predator.
</p>
<p>
The PREDATORS only die of starvation, but this happens very quickly.
If unfed, a predator will tend to starve in about 1/10 of a year.
On the other hand, the predator reproduction rate is dependent on
eating prey, and the chances of this depend on the number of available prey.
</p>
<p>
The resulting differential equations can be written:
<pre>
PREY(0) = 5000
PRED(0) = 100
d PREY / dT = 2 * PREY(T) - 0.001 * PREY(T) * PRED(T)
d PRED / dT = - 10 * PRED(T) + 0.002 * PREY(T) * PRED(T)
</pre>
Here, the initial values (5000,100) are a somewhat arbitrary starting point.
</p>
<p>
The pair of ordinary differential equations that result have an interesting
behavior. For certain choices of the interaction coefficients (such as
those given here), the populations of predator and prey will tend to
a periodic oscillation. The two populations will be out of phase; the number
of prey will rise, then after a delay, the predators will rise as the prey
begins to fall, causing the predator population to crash again.
</p>
<p>
In this program, the pair of ODE's is solved with a simple finite difference
approximation using a fixed step size. In general, this is NOT an efficient
or reliable way of solving differential equations. However, this program is
intended to illustrate the ideas of finite difference approximation.
</p>
<p>
In particular, if we choose a fixed time step size DT, then a derivative
such as dPREY/dT is approximated by:
<pre>
d PREY / dT = approximately ( PREY(T+DT) - PREY(T) ) / DT
</pre>
which means that the first differential equation can be written as
<pre>
PREY(T+DT) = PREY(T) + DT * ( 2 * PREY(T) - 0.001 * PREY(T) * PRED(T) ).
</pre>
</p>
<p>
We can rewrite the equation for PRED as well. Then, since we know the
values of PRED and PREY at time 0, we can use these finite difference
equations to estimate the values of PRED and PREY at time DT. These values
can be used to get estimates at time 2*DT, and so on. To get from time
T_START = 0 to time T_STOP = 5, we simply take STEP_NUM steps each of size
<pre>
DT = ( T_STOP - T_START ) / STEP_NUM
</pre>
</p>
<p>
Because finite differences are only an approximation to derivatives, this
process only produces estimates of the solution. And these estimates tend
to become more inaccurate for large values of time. Usually, we can reduce
this error by decreasing DT and taking more, smaller time steps.
</p>
<p>
In this example, for instance, taking just 100 steps gives nonsensical
answers. Using STEP_NUM = 1000 gives an approximate solution that seems
to have the right kind of oscillatory behavior, except that the amplitude
of the waves increases with each repetition. Using 10000 steps, the
approximation begins to become accurate enough that we can see that the
waves seem to have a fixed period and amplitude.
</p>
<h3 align = "center">
Usage:
</h3>
<p>
<blockquote>
<b>fd_predator_prey</b> <i>step_num</i>
</blockquote>
where
<ul>
<li>
<i>step_num</i> is the number of time steps to take.
</li>
</ul>
</p>
<h3 align = "center">
Licensing:
</h3>
<p>
The computer code and data files described and made available on this web page
are distributed under
<a href = "../../txt/gnu_lgpl.txt">the GNU LGPL license.</a>
</p>
<h3 align = "center">
Languages:
</h3>
<p>
<b>FD_PREDATOR_PREY</b> is available in
<a href = "../../c_src/fd_predator_prey/fd_predator_prey.html">a C version</a> and
<a href = "../../cpp_src/fd_predator_prey/fd_predator_prey.html">a C++ version</a> and
<a href = "../../f77_src/fd_predator_prey/fd_predator_prey.html">a FORTRAN77 version</a> and
<a href = "../../f_src/fd_predator_prey/fd_predator_prey.html">a FORTRAN90 version</a> and
<a href = "../../m_src/fd_predator_prey/fd_predator_prey.html">a MATLAB version</a>.
</p>
<h3 align = "center">
Related Data and Programs:
</h3>
<p>
<a href = "../../m_src/fd1d_display/fd1d_display.html">
FD1D_DISPLAY</a>,
a MATLAB program which
reads a pair of files defining a 1D finite difference model, and plots the data.
</p>
<p>
<a href = "../../f_src/fd1d_heat_explicit/fd1d_heat_explicit.html">
FD1D_HEAT_EXPLICIT</a>,
a FORTRAN90 program which
uses the finite difference method and explicit time stepping
to solve the time dependent heat equation in 1D.
</p>
<p>
<a href = "../../f_src/fd1d_predator_prey/fd1d_predator_prey.html">
FD1D_PREDATOR_PREY</a>,
a FORTRAN90 program which
implements a finite difference algorithm for predator-prey system
with spatial variation in 1D.
</p>
<p>
<a href = "../../f_src/fd2d_predator_prey/fd2d_predator_prey.html">
FD2D_PREDATOR_PREY</a>,
a FORTRAN90 program which
implements a finite difference algorithm for a predator-prey system
with spatial variation in 2D.
</p>
<p>
<a href = "../../f_src/fem1d/fem1d.html">
FEM1D</a>,
a FORTRAN90 program which
applies the finite element method to a 1D linear two point boundary value problem.
</p>
<h3 align = "center">
Reference:
</h3>
<p>
<ol>
<li>
George Lindfield, John Penny,<br>
Numerical Methods Using MATLAB,<br>
Second Edition,<br>
Prentice Hall, 1999,<br>
ISBN: 0-13-012641-1,<br>
LC: QA297.P45.
</li>
</ol>
</p>
<h3 align = "center">
Source Code:
</h3>
<p>
<ul>
<li>
<a href = "fd_predator_prey.f90">fd_predator_prey.f90</a>, the source code.
</li>
<li>
<a href = "fd_predator_prey.sh">fd_predator_prey.sh</a>,
commands to compile the source code.
</li>
</ul>
</p>
<h3 align = "center">
Examples and Tests:
</h3>
<p>
The program writes out a file of the solution data. The data can be
plotted by MATLAB, for instance, using commands like this:
<pre>
trf = load ( 'trf_10000.txt' );
plot ( trf(:,1), trf(:,2), 'g-', trf(:,1), trf(:,3), 'r-' )
title ( 'A Predator Prey System' );
xlabel ( 'Time' );
ylabel ( 'Population' );
</pre>
</p>
<p>
<b>TRF_100</b> uses 100 timesteps, which are not enough.
<ul>
<li>
<a href = "trf_100.txt">trf_100.txt</a>, the data.
</li>
<li>
<a href = "trf_100.png">trf_100.png</a>, a plot.
</li>
</ul>
</p>
<p>
<b>TRF_1000</b> uses 1000 timesteps; the solution does not explode,
and seems to show periodicity, except that it is clearly growing.
<ul>
<li>
<a href = "trf_1000.txt">trf_1000.txt</a>,
a table of the prey and predator values using 1000 steps.
</li>
<li>
<a href = "trf_1000.png">trf_1000.png</a>,
a plot.
</li>
</ul>
</p>
<p>
<b>TRF_10000</b> uses 10000 timesteps. The cyclic nature of
the solution is clear.
<ul>
<li>
<a href = "trf_10000.txt">trf_10000.txt</a>,
a table of the prey and predator values using 10000 steps.
</li>
<li>
<a href = "trf_10000.png">trf_10000.png</a>,
a plot.
</li>
</ul>
</p>
<h3 align = "center">
List of Routines:
</h3>
<p>
<ul>
<li>
<b>FD_PREDATOR_PREY</b> solves a pair of predator-prey ODE's.
</li>
<li>
<b>GET_UNIT</b> returns a free FORTRAN unit number.
</li>
<li>
<b>R8MAT_WRITE</b> writes an R8MAT file.
</li>
<li>
<b>S_TO_I4</b> reads an integer value from a string.
</li>
<li>
<b>S_BLANK_DELETE</b> removes blanks from a string, left justifying the remainder.
</li>
<li>
<b>TIMESTAMP</b> prints the current YMDHMS date as a time stamp.
</li>
</ul>
</p>
<p>
You can go up one level to <a href = "../f_src.html">
the FORTRAN90 source codes</a>.
</p>
<hr>
<i>
Last revised on 28 June 2012.
</i>
<!-- John Burkardt -->
</body>
</html>