Experiences with Computer Simulations of Primate Populations
Green Creek Paradigms, LLC
For decades, computer simulations have held advantages over mathematical models of primate populations. Primates are highly social animals, often living in groups with complex, hierarchical, dynamic structures. However, particulars of primate group structures and dynamics vary considerably among species. Age, sex, and kinship relationships often provide bases for behavioral organizational features seen in primate groups. As a result, numerous interconnections appear to exist between demographic, genetic, and behavioral processes in primate populations. The complexity known to exist in primate populations is difficult to model mathematically. However, simulations that model individual life histories and use syntax-like data structures borrowed from abstract language theory have proven adaptable and effective.
Several generations of primate population simulation libraries using these approaches have been written in FORTH, C, and Python languages. The current Python language system, named CRITTRZ, makes heavy use of object-oriented architecture and incorporates a run-time interface to a geographic information system, for representation of landscapes occupied by simulated populations. This system is designed to support basic scientific research and planning decisions by conservation biologists. Recent applications of the current system have allowed a detailed examination of interplays between demographic circumstances of simulated populations, group subdivision processes, and gene distributions among groups. This system adds support for modeling infectious disease transmission and multiple species in the same environment. This work has required extended shifts of focus between considering results from field studies, contemplating formal models that would best portray events seen in the field, implementing software, and conducting simulations.
The author has found that results of primate population simulations often include unanticipated findings that lead to further investigations and give the process an ongoing exploratory character. One goal of the author is to apply approaches developed for modeling primate populations to populations of nonprimate species. It will be interesting to see to what extent formal structures that have been useful in representing primate group organizations remain useful when applied to populations of other kinds of mammals.