The precision and accuracy of the Random Encounter and Staying Time Model’s estimation of species population density

Hogg, Jennifer (2022) The precision and accuracy of the Random Encounter and Staying Time Model’s estimation of species population density. Masters thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (1MB)

Abstract

1 Introduction: Species density is perhaps the most sought-after measurement in ecological research because it has a key role in conservation management practices and species monitoring. One method to measure density is to implement camera traps in ecological environments that takes continuous photographs at short time intervals to create a timelapse, or records a video of animals throughout the night and day. Camera-trap data can be used to derive density estimates using the Random Encounter and Staying Time (REST) model for non-distinguishable individuals in a population. 2 Methods: I mathematically recreate the REST model under the theoretical framework of ideal gas law physics. I use this as a basis to derive the mean and variance of the REST model using probability density functions and mathematical moments. I use three different detection zone areas, research periods, and animal speeds to see how it affects the accuracy and precision of the density estimates. 3 Results: Assuming all assumptions of my model have been met, the REST model will give biased density estimates depending on the detection zone shape and the movement patterns of the species. The model’s density estimates become more precise for longer research periods and larger detection zones. Faster moving animals also produce more precise density estimates. The mean estimate remains a true reflection of the species density regardless of camera detection zone, research period, or animal speed. 4 Synthesis and application: My work uses a combination of statistical distributions and mathematics to predict pre-emptively the precision and accuracy of the REST model without empirical data. This allows researchers to be able to change the REST model’s parameters, research period and detection zone area, in accordance with the species movement speeds to have an idea about the expected results the REST model will provide. Given that our work relies strictly on theoretically reasoning, we believe that this allows for our work to be applicable to a broad range of species, compared to if we had used empirical evidence. Given the popularity of the REST model, our work is anticipated to be very relevant to many future research monitoring projects.

Actions (login required)

View Item