Population projection models in their many forms are not new – they are key components of any effort to manage a species or population. The ability to make formal predictions about the probable effects of management choices is crucial when making population management decisions. While it is impossible to measure extinction probability solely with field data, predictive models can be used to develop alternative metrics that represent extinction risk. This is where recovery criteria come in, as they are defined for each studied population and used as measurable attributes of extinction probability. In this study, researchers proposed that simulation models and statistical analysis of the simulated data can be used as the link between measurable attributes of a population and extinction probability.

To accomplish this, the study used mathematical simulation of extinction probabilities of piping plovers, a small shorebird, in the Great Plains. The scientists claim that their method provides a straightforward way of developing specific and measurable recovery criteria (e.g. population abundance, reproductive rate, etc.) linked directly to extinction risk.

The researchers warn that the model must be able to account for imperfect field data. Whether due to flawed detection or sampling variation, uncertainty is nearly unavoidable. Inaccurate measurements can affect the results of the extinction probability predictions and potentially lead to premature or delayed delisting of the species in question. The researchers propose either using only metrics that can be accurately measured or by including a safety factor of sorts when setting up the recovery criteria. In this model, developed for piping plovers, the team decided to keep survival rates, a demographic parameter, consistent among regions, while some other parameters were based on limited data. Such assumptions are not uncommon in mathematical modelling and must therefore be kept in mind. The more accurate the data, the more thorough the model.

The model they created was used to predict extinction risks region by region and for the whole population of piping plovers for 50 years, assuming that current demographics and management efforts continue for that time. The Southern Rivers region, where intensive habitat management has occurred in recent years, had the lowest extinction probability and seemed to significantly lower the extinction risk of the overall population. Meanwhile, it was found that extinction probabilities in the other three studied regions (i.e. Northern Rivers, alkali wetlands and Prairie Canada) exceeded 6% at 50 years.

The model is supposed to be universal – it and the subsequent analysis of the data make it possible to quantify the extinction probability for any given starting population size and expected population growth rate for the modelled species. Potentially, this type of analysis could be done for any measurable population parameter. Using the proposed approach, a species/population recovery team could select an acceptable extinction risk for the protected species and afterwards use the data analysis to identify factors that would lead to recovery for the given population. The recovery team could then set delisting criteria and manage the population accordingly to reducing extinction risk.

Animal advocates may use the findings of this study to promote advancements in species conservation efforts. The advent of advanced mathematical modelling of populations is bound to raise awareness on more threatened species, and instigate action sooner. With this tool in the toolbox, scientists could reduce risk of extinction more efficiently.