Extinction risk depends strongly on factors contributing to stochasticity

Area of Science:

• Ecology
• Population Dynamics
• Conservation Biology

Background:

• Extinction risk in natural populations is influenced by stochastic factors.
• Stochasticity encompasses demographic stochasticity, environmental stochasticity, and demographic heterogeneity.
• Previous models often failed to integrate all stochasticity types for comprehensive risk assessment.

Purpose of the Study:

• To develop and analyze mechanistic stochastic Ricker models incorporating demographic stochasticity, environmental stochasticity, and demographic heterogeneity.
• To investigate the combined effects of these stochastic factors on population extinction risk.
• To compare the predictive power of a full stochastic model against simpler, conventional models.

Main Methods:

• Derivation of a family of stochastic Ricker models with varying combinations of stochastic factors.
• Application of the full stochastic model to a laboratory population of Tribolium castaneum.
• Comparative analysis of model outputs to determine the relative importance of different stochasticity sources.

Main Results:

• Extinction risk is highly dependent on the specific combination of stochastic factors included in the model.
• The full stochastic model is necessary for accurately determining the relative importance of environmental versus demographic variability.
• In Tribolium castaneum, demographic stochasticity was the primary driver of variability, contrary to conclusions from simpler models.

Conclusions:

• Current estimates of extinction risk may be significantly underestimated due to misattribution of variability to environmental factors instead of demographic ones.
• Demographic sources of stochasticity entail a higher extinction risk for equivalent variability levels compared to environmental sources.
• Accurate extinction risk assessment necessitates the use of comprehensive stochastic models that include all relevant demographic and environmental factors.