This kind of studies are hard to conduct since there are so little empirical data, hence the use of theorethical modeling.

The method used was mathematical modeling based on Lokta-Volterra equations with a functional response of type II performed in the program Matlab.

The analyses based on Christianou & Ebenmans (2006) deterministic food-web models which they used for reintroducing species into communities. The model community that was generated is a triangular shaped deterministic food-web with 12 species divided into three trophic levels; five basal species (primary producers), four intermediate species (links both up and down in the food-web) and three top predator species (only links down in the food-web).

Each species was removed from the food-web one at a time. When the species was removed from the ecosystem its attributes (mortality- or growth-rate, depending on if the species is a consumer or a producer) was changed with 40% and 75%, as if they were kept in captivity during their absence from the ecosystem. The mortality/growth-rate was used since it is a good measure of several changes in a captive bred species, for example, genetic depression, weakened predator skills and infertility all affect these rates. Different mortality-/growth-rates were used due to the fact that captivity alters the mortality of species in different extent and can also be illustrated as a shorter/ longer time spent in captivity. As a control-sample the species was removed and reintroduced without any change in its mortality/growth-rates (0% change). When the species was removed and then reintroduced to the food-web, the effects of this was assessed and statistically analyzed.