Using uncertainty estimates in analyses of population time series

Area of Science:

• Ecology
• Population Dynamics
• Conservation Biology

Background:

• Wildlife monitoring provides essential data for conservation, often as time series of population abundance estimates.
• Analyzing population dynamics is limited by the lack of sampling error information in abundance estimates.
• Previous methods using replicated sampling are effective but can be complex to implement.

Purpose of the Study:

• To evaluate a method that incorporates standard errors with population estimates in state-space models.
• To account for sampling variability in population dynamics analyses.
• To provide a simpler, effective approach for wildlife monitoring data.

Main Methods:

• Simulated data from a Gaussian state-space model with multiple observations per time point.
• Incorporated standard errors with population estimates to model sampling variability.
• Tested the method with heteroscedastic observation error, site effects, and correlated observations.

Main Results:

• The method using standard errors performed similarly or better than using all raw observations.
• Effectiveness was maintained even with few observations per time point.
• The approach showed good performance on real data from the North American Breeding Bird Survey.

Conclusions:

• Using standard errors with population estimates is a simple and effective method for analyzing population dynamics.
• This approach enhances wildlife monitoring and conservation by accounting for sampling variability.
• The method is adaptable to various sampling procedures and data types.