Assessment of Sea-Turtle Status and Trends: Integrating Demography and Abundance (2010)

The island model is a basic model of division of a species into isolated “islands”. The organisms breed randomly within an island, and migrants are drawn randomly from each island. Because of the division and isolation, the heterozygosity (when two different alleles occupy the gene’s positions, or loci, on a pair of chromosomes) of the entire group is lower than would be expected if there were random mating among all members of the species (Wright, 1943).

When a species inhabits a large geographic area, genotype frequencies may change gradually in space in a way that is not due to physical barriers. Such isolation occurs because the geographic area is much greater than an individual’s migration distance. This model reflects the loss of heterogeneity that results when individuals breed with their neighbors and produce genetic differentiation across the range. At the ends of the range, there is smaller genetics correlation than in nearby localities (Wright, 1943). Dethmers et al. (2006) observed isolation by distance—greater than 2,000 km—in west Pacific green turtles (Chelonia mydas).

This model “assumes that the entire population is subdivided into colonies and the migration of individuals in each generation is restricted to nearby colonies” (Kimura and Weiss, 1964). Thus it is a special case of isolation by distance.

Sewall Wright and others developed the ideas of spatial structure in populations relative to their genetics. Levins (1969, 1970) reframed the effects of spatial structure to the population dynamics and ecology of a species. His model concentrated on the effects of extinction and recolonization of local populations on the persistence of a species. Hanski and Simberloff (1997) have developed the ideas further since the 1980s to emphasize the effect of migration and connectivity on the vital rates of local populations and how spatial heterogeneity can act to protect a species from extinction. Modern theory does not necessarily assume that local populations will go extinct and allows that there can be substantial migration between them. However, the ramifications of habitat fragmentation on formation of metapopulations has not been fully developed (Jones, 2006).