Extending density surface models to include multiple and double-observer survey data.

Area of Science:

• Ecology and Wildlife Management
• Spatial Modeling
• Distance Sampling

Background:

• Spatial models are crucial for ecological research (habitat use) and wildlife management (population monitoring).
• Integrating data from multiple, diverse observation processes into spatial models presents a significant challenge.
• Existing Density Surface Models (DSMs) do not adequately handle multiple data sources.

Purpose of the Study:

• To extend Density Surface Models (DSMs) to integrate data from multiple distance sampling survey types.
• To develop a unified framework for analyzing diverse distance sampling data within a single spatial model.
• To improve the robustness and accuracy of abundance and density estimations by incorporating multiple data sources.

Main Methods:

• Extended the Density Surface Model (DSM) framework to accommodate multiple data sources.
• Integrated data from conventional distance sampling, double-observer distance sampling, and strip transects.
• Employed variance propagation to ensure accurate uncertainty quantification in abundance estimates.
• Implemented the methodology in the `dsm` R package.

Main Results:

• Successfully demonstrated the integration of multiple distance sampling survey types into a single spatial model.
• The extended DSM approach provides more robust abundance and density estimates compared to single-source methods.
• Variance propagation effectively accounts for uncertainty from different data sources.

Conclusions:

• The novel methodology enables the simultaneous analysis of diverse distance sampling data within a unified spatial model.
• This approach leads to more reliable and comprehensive wildlife abundance and density estimations.
• The developed methods are applicable to broader geographical and temporal scales, enhancing ecological research and wildlife management.