Detecting mass mortality events in wildlife populations

Area of Science:

• Ecology and Wildlife Health
• Conservation Biology
• Epidemiology

Background:

• Increasing reports of mass mortality events (MMEs) across diverse taxa.
• Significant underestimation of MMEs due to detection biases (e.g., infrequent sampling, cryptic species, rapid carcass decay).
• Challenges in assessing temporal and spatial trends of MMEs.

Purpose of the Study:

• To develop a modeling framework to understand factors influencing MME detection probability.
• To provide insights into optimal sampling frequencies and detectability dynamics for surveillance.
• To estimate MME frequency in a specific endangered salamander population.

Main Methods:

• Developed a probabilistic model linking MME detection probability to observation frequency and detectability decay rates.
• Applied the framework to analyze Ranavirus-related MMEs in 35 populations of an endangered salamander subspecies.
• Evaluated the impact of sampling frequency and detectability on the likelihood of observing an MME.

Main Results:

• The probability of detecting an MME in the studied salamander populations was less than 50%.
• The actual frequency of MMEs in the system is likely higher than the single observed event.
• Model demonstrates that higher peak detectability, slower decay rates, and increased observation frequency enhance MME detection probability.

Conclusions:

• Current surveillance likely underestimates the true incidence of MMEs, particularly for cryptic species or those with rapid decay.
• The developed framework aids in designing more effective wildlife disease surveillance programs.
• Further research is needed to address limitations such as assumptions of regular, equal-effort surveys.