Research Document - 2011/118
Biological Sampling Of BC Herring: Analysis Of Sampling Requirements For Characterizing Age Structure And Other Biological Characteristics Of Fisheries And Spawning Populations
By V. Haist
Data collected through the herring biological sampling program is one of the major inputs to the stock assessment models, but a comprehensive review of the program and data collected through the program has not been completed. The work presented here, a continuation from an initial review of the biological sampling program presented to CSAP in September 2010, focuses on analyses of samples from the roe and test fisheries collected since 1972. Specific questions addressed in the report are:
- Is there evidence for spatial or temporal structure (sub-stocks) within herring stock assessment regions?
- Do Sn-roe and test fisheries sample different populations?
- Are there sex-related differences in biological characteristics that should be captured in the stock assessment?
- Can gonadosomatic indices be used to associate herring samples with spawning events?
- Is the assumption that test fishery samples are representative of spawning populations (in some years) reasonable?
- Is there an objective basis for weighting roe fishery and test fishery samples in developing age-compositions for stock assessments?
- Has the precision of age compositions changed over time?
- How would a decrease in sampling effect the precision (and accuracy) of age-compositions?
Cluster analysis of seine-caught age compositions indicates that samples taken within each stock assessment region are unlikely to come from a single homogeneous population. There is a high degree of consistency in the age composition of samples taken with most sections, and consistency among samples decreases with increased geographical separation. On average, the mean age of samples taken from seine roe fisheries tend to be slightly higher than those taken in the test fishery, and the mean age of samples taken early in the spawning season tend to be slightly higher than the mean age of samples taken later in the spawning season.
Gillnet roe fisheries are highly selective for female fish at younger ages (to age 6 or 7), while sex ratios in seine-caught samples indicate a higher proportion of males (than females) for ages 2 and 3. Sex-specific differences in length-at-age are trivial.
Gonadosomatic indices (ratio of ovary weight to body weight) show promise for predicting when sampled herring would reach full maturity, and hence potentially associating samples with spawning events. Further work is required to develop a methodology.
There are no indications that the precision of age compositions (seine and gillnet roe fisheries and test fisheries) have decreased in recent years. Age composition estimates from the seine and gillnet roe fisheries are generally quite precise, and a reduction in the number of samples collected from these fisheries would not likely compromise the integrity of the data for stock assessments. Collection of 6 to 12 samples from each fishery would result in “effective” sample sizes of approximately 300 fish, resulting in reasonable c.v.s for proportions-at-age that are not small (c.v.s ≤ 0.25 for proportions ≥0.05). For the test fishery, a reduction in sampling effort (simulated by 14 day sampling periods) would potentially decrease the precision and accuracy of age composition estimates substantially.
This document is available in PDF format. If the document is not accessible to you, please contact the Secretariat to obtain another appropriate format, such as regular print, large print, Braille or audio version.
- Date modified: