Exploring the effects of methodological choices on the estimation and biological interpretation of life history parameters for harbour porpoises in Norway and beyond.

• Published in: PLOS ONE
• Main Authors: Anne Kirstine Frie, Ulf Lindström
• Format: Article in Journal/Newspaper
• Language: English
• Published: Public Library of Science (PLoS) 2024
• Subjects: Medicine; R; Science; Q; greenlandic; Harbour porpoise
• Online Access: https://doi.org/10.1371/journal.pone.0301427; https://doaj.org/article/7aef065cfeee450fa904a6ba32d2f468

This study investigates effects of subtle methodological choices on the estimation and biological interpretation of age, growth and reproductive parameters for harbour porpoises. The core analyses are based on a focal Norwegian data set built on samples from 134 harbour porpoises caught incidentally in gillnet fisheries along the Norwegian coast during autumn 2016 and spring 2017. Two contrasting practices for interpretation of seasonal and ontogenetic characteristics of tooth growth layer formation resulted in significant age differences among spring samples of young porpoises and for older animals across seasons. In turn, these differences affected estimates of age at maturity and asymptotic lengths, respectively. We also found significant differences in male age at maturity between two well-documented maturity criteria and between mathematical estimators of age at maturity for both sexes. Two different criteria for corpus albicans classification furthermore resulted in different patterns of ovarian corpora accumulation, which may affect some estimates of fecundity rates and contaminant loads. Both corpora accumulation patterns were also found in reanalysed data from German and Greenlandic porpoises. Based on tabulated overviews of methodological choices made in previous harbour porpoise studies, we argue that several of the issues mentioned above have wider relevance and may affect the validity of meta-analyses as a tool for estimating harbour porpoise sensitivity to extrinsic pressures. Differences in cause of death (COD) composition between data sets can have a similar effect. We demonstrate this in a meta-analysis of published harbour porpoise pregnancy rates, showing significantly higher values for trauma-killed samples compared to samples comprising mixed COD categories. COD also affected the estimated impacts of three previously analysed extrinsic predictors as well as an added predictor for vessel noise levels. We discuss the potential contributions of methodological, biological and anthropogenic ...