A method of analyzing density-dependent vital rates with an application to the Gainj of Papua New Guinea.

Abstract

A method of estimating age-specific coefficients of density-dependent variation in fertility and mortality is developed; the method is applicable to longitudinal data on population size and the number of births and deaths classified by age. Given a sufficiently large data set, it is possible to estimate both the sensitivity of each age class to density-dependent damping and the density effect of each age class on every age class in the population. Application of the method to government census data on the Gainj, a small tribal population from highland Papua New Guinea, shows that fertility is density-independent, but that mortality is at least partially density-dependent. This finding suggests that the size of the population is regulated by mortality rather than fertility. Individuals aged less than five years and greater than 50 years are particularly sensitive to density-dependent survival damping; individuals of adolescent and early reproductive age are not themselves damped, but appear to be responsible for the observed damping.