Melding wildlife surveys to improve conservation inference

Area of Science:

• Ecology
• Conservation Biology
• Statistical Modeling

Background:

• Species of conservation concern are often monitored by multiple entities, generating diverse datasets.
• Individual datasets may have limitations like low resolution, biased sampling, or high uncertainty, hindering effective management.
• Integrated models offer a framework to combine these datasets, overcoming individual data deficiencies.

Purpose of the Study:

• To develop an integrated modeling approach for assimilating disparate ecological survey data.
• To provide shared inference on spatio-temporal trends for species of conservation concern.
• To improve predictive performance and inform management decisions for lesser prairie-chicken populations.

Main Methods:

• Developed an integrated modeling approach using Bayesian Markov melding.
• Assimilated density estimates from independent aerial and ground surveys of lesser prairie-chicken.
• Applied data augmentation strategies for efficient sampling and analysis.

Main Results:

• The integrated model demonstrated improved predictive performance compared to analyzing surveys independently.
• The model successfully assimilated data with distinct sources of observational error.
• Enabled prediction of lesser prairie-chicken density in unsampled areas.

Conclusions:

• Integrated models are effective for combining diverse ecological datasets, even with varying spatio-temporal resolutions and uncertainties.
• This approach enhances the analysis of species of conservation concern, providing more robust insights for management.
• The study highlights the value of integrated modeling for improving conservation outcomes and understanding population dynamics.