Saturday, July 19, 2014

Simulating outcomes

Here are some simulations inspired by original question and hypotheses submitted in the second part of the assignment.
I selected a 5 min count of arrivals at grass border obtained on June 23 (central trail, evening). Using the excel random function, I added to these real field data simulated crossing times t and simulated covered distance during crossing d (two variables to be measured on video recordings of the trail traffic in the study project). Covered distance increases by zigzagging and represents an excess of work.
Here are simulated possible outcomes if the following hypotheses are true:
1. Traffic in bursts doesn't save time or energy (light blue curves). 
2. Help from escorting leaders acts to save time and energy (red curves)
3. Enhanced marking by trail laying acts to save time and energy (dark blue curves).
Hypotheses 1 and 2 are mutually exclusive. So are hypotheses 1 and 3. Hypotheses 2 and 3 can coexist.



The principal inspiration came from an ant colony in my backyard. I however acknowledge the fantastic published works of the "ant team" of the Université libre de Bruxelles (ULB). In their abstracts and papers with free web access, I could find the confirmation that my project probably stands on sufficiently solid scientific background. It also helped me in the wording of the project. I didn't contact in person the people from the ULB, because I am not a biologist and am retreated from an other Faculty.






No comments: