Parameters and model design for previously created army ant models:
Movement matrix for Deneubourg et al., 1989: Movement Matrix, Solé et al., 2000:
 |
 |
||
Description
|
Parameter |
Value |
Notes
|
|
|
Deneubourg et al., 1989 |
Sole* |
|||
|
Quantity
of pher. laid per step |
P |
Out: 1 Returning: 10 |
47 |
|
|
Pheromone
concentration at point (i,j) |
f
(i,j) |
varies |
|
|
|
Concentration
of pheromone with ˝ prob. of movement |
P ˝ |
100 |
100 |
|
|
Pheromone
Evap. rate |
e | d |
1/30 |
2.1 x 10 -3 |
|
|
Attractiveness
of empty nodes |
m |
5 |
47 |
|
|
Cut-off
for dropping pheromone – going out |
Psat-in |
1000 |
51 |
|
|
Cut-off
for dropping pheromone – going in |
Psat-in |
300 |
390 |
Pher. drop parameters were chosen from previous
work on other ant species (Den. et al. 1989, “Self-organizing
exploratory pattern of the Argentine Ant”). |
|
Probability
of movement |
p
L, O, R |
see Equations
section |
|
|
|
Max
num of ants / patch |
Am |
20 |
30 |
|
|
Length
of model run (time-steps) |
|
300, 600, 900 |
1000 |
|
|
Number
of ants leaving nest per time-step |
|
10 |
10 |
Ants
don’t leave if next point is full |
|
Grid
shape / # of nodes to move to |
|
diamond / 2 |
square / 3 |
|
|
|
|
|
|
|
|
|
|
|
|
*
Values for Sole et al are the optimized values for E. burchelli-like
food distribution. |
Equations
|
|
|
|
|
|
|
|
Parameter |
Value |
Description |
|
Pm |
|
Probability
of movement |
|
pL |
|
Probability
of choosing node A (Northwest node) |
|
pR |
|
Probability
of choosing node C (Northeast node) |
|
pO |
1
- pL - pR |
Probability
of choosing node B (North node) |
|
C |
|
|
|
|
|
|
|
|
|
|
