Are there biases in biopsy sampling? Potential drivers of sex ratio in projectile biopsy samples from two small delphinids
Article first published online: 5 FEB 2013
2013 by the Society for Marine Mammalogy Published 2013. This article is a US Government work and is in the public domain in the USA.
Marine Mammal Science
Volume 29, Issue 4, pages E366–E389, October 2013
How to Cite
Kellar, N. M., Trego, M. L., Chivers, S. J., Archer, F. I., Minich, J. J. and Perryman, W. L. (2013), Are there biases in biopsy sampling? Potential drivers of sex ratio in projectile biopsy samples from two small delphinids. Marine Mammal Science, 29: E366–E389. doi: 10.1111/mms.12014
- Issue published online: 9 OCT 2013
- Article first published online: 5 FEB 2013
- Manuscript Accepted: 30 AUG 2012
- Manuscript Received: 12 JAN 2012
- sex ratio;
- sampling bias;
- Delphinus capensis ;
- Delphinus delphis
Molecular assays were used to determine the sex of 1,294 biopsied common dolphins (658 long-beaked common dolphins, Delphinus capensis, and 636 short-beaked common dolphins, D. delphis) in the Southern California Bight. Sex ratio differed substantially between the two species; females comprised 241 (36.6%) of D. capensis samples and 410 (64.5%) of D. delphis samples. All biopsies were taken either from a large research ship or from a small, rigid-hull inflatable boat (RHIB) launched from the larger ship. When conducting replicate biopsy effort on the same schools from each vessel/platform (“Tandem Biopsy Sampling”), we found evidence that disproportionately more female D. capensis were biopsied from the RHIB than from the ship but the same was not true for D. delphis. We suspect that these results are driven by bowriding-behavior differences between the two species. Biopsy duration, geographic location, school size, and Julian date were considered as potential covariates with sex ratio; geographic location was the only one to show strong evidence of correlation. This study also presents an alternative to the erroneous practice of comparing sex ratios to a theoretical assumption of parity (i.e., 50:50 sex ratio) when researchers avoid sampling animals paired with calves.