Over this past summer (2010) I studied lateralization of Caspian Terns with Willamette University's David Craig. In the process of collecting data to study lateralization, we ended up with 200+ hours of video of Caspian Terns feeding their young. We also have video of Tern chicks being picked off (predated) by gulls or harassed and sometimes killed by other Terns. Of course, some of the Tern chicks we video taped did make it long enough to fledge.
My goal for my senior computer science project is to develop a biological model and simulation to predict chick survival rates given colony conditions such as gull population, fish availability, parent protectiveness, etc. I will need to determine a type of model that will work well for this purpose and then learn how to apply it to my project. I will need to develop the model as completely as I can and write a simulation environment that variables can be entered into to receive an estimated survival rate in return.
Check this out:
ReplyDeletehttp://www.red3d.com/cwr/ibm.html
I think the IBM is relevant to your work. -hy
Thanks. IBM is definitely something I am looking into, especially since I have always had an interest in agent-based modeling. I am not sure yet if that is the approach that will win out over mathematical modeling of the entire population as one entity, but I think it is likely to provide more useful results.
ReplyDeleteBayesian Animal Survival Estimation
ReplyDeleteS. P. Brooks, E. A. Catchpole and B. J. T. Morgan
http://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.ss/1009213003