Effects of model formulation on estimates of health in individual right whales (Eubalaena glacialis)

Robert S. Schick*, Scott D. Kraus, Rosalind M. Roll, Amy R. Knowlton, Philip K. Hamilton, Heather M. Pettis, Len Thomas, John Harwood, James S. Clark

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Right whales are vulnerable to many sources of anthropogenic disturbance including ship strikes, entanglement with fishing gear, and anthropogenic noise. The effect of these factors on individual health is unclear. A statistical model using photographic evidence of health was recently built to infer the true or hidden health of individual right whales. However, two important prior assumptions about the role of missing data and unexplained variance on the estimates were not previously assessed. Here we tested these factors by varying prior assumptions and model formulation. We found sensitivity to each assumption and used the output to make guidelines on future model formulation.

Original languageEnglish
Title of host publicationAdvances in Experimental Medicine and Biology
PublisherSpringer
Pages977-985
Number of pages9
DOIs
Publication statusPublished - 2016

Publication series

NameAdvances in Experimental Medicine and Biology
Volume875
ISSN (Print)0065-2598
ISSN (Electronic)2214-8019

Keywords

  • Bayesian
  • Body condition
  • Missing data
  • Photo identification
  • Process error

Fingerprint

Dive into the research topics of 'Effects of model formulation on estimates of health in individual right whales (Eubalaena glacialis)'. Together they form a unique fingerprint.

Cite this