How to infer population trends in sparse data: examples with opportunistic sighting records for great white sharks

Abstract

AbstractAim Biodiversity is declining at accelerating rates. Understanding past and ongoing changes to biodiversity is paramount in prioritizing conservation action and restoring functional ecosystems. Yet long‐term, systematic data on the distribution and abundance of species are sparse. For many organisms, specimen collections and anecdotal accounts of chance sightings or captures constitute the only source of information. Such data have the potential to provide valuable insights on long‐term ecosystem changes, but are often neglected because they are difficult to analyse quantitatively. Here we review available methods and introduce a new approach.Location Historic data on sightings and captures of great white sharks in the eastern Adriatic and off eastern Canada serve to illustrate the utility of both the existing methods and the new approach.Method Unlike existing methods, the new approach focuses on estimating population trends rather than verifying extinction and explicitly addresses uncertainty over observation effort via two tiers of sensitivity analysis. It fits a series of generalized linear models that provide multiple estimates of declines under alternate scenarios regarding the appropriate reference period and observer trends. Programming code to implement the approach in freely available software is provided as supplementary material.Results Example analyses of great white shark sightings suggest that local populations of this species have suffered dramatic declines, both in the eastern Adriatic and along Canada’s eastern coast. Although not yet extinct, this top predator may therefore no longer be able to fulfil its former ecological role.Main conclusions Careful quantitative analyses of imperfect historical data can provide valuable insights into past ecological changes. Such insights are crucial to improved management and restoration of individual species and their ecosystems. We therefore hope that our review of available methods will facilitate quantitative evaluations of species for which analysis was previously impeded by a lack of systematically collected data.