A generalised random encounter model for estimating animal density with remote sensor data | JoVE Visualize

Area of Science:

Wildlife ecology and conservation technology
Ecological modeling and population estimation

Background:

Remote sensing technologies like camera traps and acoustic detectors are increasingly used for wildlife monitoring in diverse environments.
Existing methods for estimating animal abundance or density often require individual recognition or precise distance measurements, which are challenging to obtain.
The standard random encounter model (REM) is effective for camera trap data but not acoustic data due to signal directionality.

Purpose of the Study:

To develop and validate a generalized random encounter model (gREM) applicable to both camera trap and acoustic detector data.
To assess the accuracy and precision of the gREM across various sensor and animal signal widths, capture efforts, and movement patterns.

Main Methods:

Derived a generalized random encounter model (gREM) accounting for sensor detection widths and animal signal directionality.
Utilized simulations to test the gREM's performance with different parameter combinations (sensor/signal widths, capture numbers, movement models).
Evaluated model accuracy and precision based on simulation outputs.

Main Results:

The gREM accurately estimates absolute animal density across all tested combinations of sensor detection and animal signal widths.
Model precision improves with larger sensor detection and animal signal widths, and with increased numbers of captures.
Animal movement models did not significantly affect the accuracy or precision of the gREM estimates.

Conclusions:

The gREM is a robust and effective method for estimating absolute animal densities from remote sensor count data in both terrestrial and marine environments.
The model is versatile, applicable to data from camera traps and acoustic detectors for various species (e.g., mammals, birds, cetaceans, bats).
The gREM offers a valuable tool for monitoring unmarked wildlife populations across broad scales as remote sensing technology becomes more widespread.