I have gone through the first 20 or so sources that made it past my initial screening for interest and narrowed it down to 5 sources that have real potential for being of use to me in my study. They are cited below along with justifications for my interest in them.
Galla, T. (2010). Independence and interdependence in the nest-site choice by honeybee swarms: Agent-based models, analytical approaches and pattern formation. Journal of theoretical biology, 262(1), 186-96. Elsevier. doi: 10.1016/j.jtbi.2009.09.007.
In this paper a model that predicts the necessary levels of independence and interdependence for bees to successfully pick the best new hive site is explained. The model presented is largely solvable, in its simplified form, without computers. One nice thing about the model is its ability to represent the interplay of two opposing strategies for hive selection.
Bithell, M., Brasington, J., & Richards, K. (2008). Discrete-element, individual-based and agent-based models: Tools for interdisciplinary enquiry in geography? Geoforum, 39(2), 625-642. doi: 10.1016/j.geoforum.2006.10.014.
This paper is a discussion of the differences and merits of three different types of discrete models and it gives examples of each in the context of geography. Discrete-element models (DEM) work with a finite number of differentiable elements. The elements are fairly simple in and of themselves, it is the interactions that are of interest. Individual-based models (IBM) are used when the internal dynamics of individual entities in the simulation are important or intricate. Agent-based models (ABM) are used when entities not only have intricate internal dynamics but also decision making capabilities. Based on these descriptions, my model will either be an IBM if I do not attempt to make my model behavioral or an ABM if I want to model behavior of the terns as decision processes.
Harel, D., Setty, Y., Efroni, S., Swerdlin, N., & Cohen, I. R. (2008). Concurrency in biological modeling: Behavior, execution and visualization. Electronic Notes in Theoretical Computer Science, 194(3), 119-131. doi: 10.1016/j.entcs.2007.12.009.
In this paper the authors describe their method for simplifying a model: split it into seperate smaller models work on them independently. Concurrency is also a big topic in this paper. That is, the complications that arise from natural processes being largely concurrent in real life and the benefits of being able to run a model in segments as well as concurrently.
Morgan, B. J., Catchpole, E. a., & Brooks, S. P. (2000). Bayesian Animal Survival Estimation. Statistical Science, 15(4), 357-376. doi: 10.1214/ss/1009213003.
The authors discuss a method for modeling survival, particularly of birds. The title implies a perfect fit for my project, but the examples given deal with year to year survival of birds with model parameters such as recapture or re-sighting probability. I may find a way to adapt this to the my needs (modeling survival of chicks within a breeding season). One problem with the model is it predicts survivability based on a probability of resighting, which is likely to be based on multiple years of data. Although I could probably come up with resighting data, the model would not be based upon annual differences in gull predation or fish availability.
Pollock, K. H., & Cornelius, W. L. (1988). A Distribution-Free Nest Survival Model. Biometrics, 44(2), 397. doi: 10.2307/2531854.
This paper could be very helpful if I decide to base my model on the methods described within. The title of the article says it all: modeling survival of young birds during nesting, which is exactly what I plan to do. The only assumption made in the paper that is not relavent is that we may not have found all of the nests we are studying. In reality, the nature of East Sand Island allows us to be certain that all nest scrapes the terns nested in were in view from at least one of the blinds. However, his seems like a simple enough detail to work around in adapting a model.
No comments:
Post a Comment