Migration as a source of bias in the measurement of cephalopod growth

• Published in: Antarctic Science
• Main Authors: Hatfield, Emma M.C., Rodhouse, Paul G.
• Format: Article in Journal/Newspaper
• Language: English
• Published: Cambridge University Press (CUP) 1994
• Subjects: Geology; Ecology, Evolution, Behavior and Systematics; Oceanography; Antarctic Science
• Online Access: http://dx.doi.org/10.1017/s0954102094000283; https://www.cambridge.org/core/services/aop-cambridge-core/content/view/S0954102094000283

Theory predicts that, in a closed exploited population showing no sampling or other bias, the mean age of fish or squid between one sampling date and the next should increase by the time interval between samples. Age data, derived from statoliths, have been used to test the hypothesis that the effect of migration through an area where a population of Loligo gahi is sampled would be apparent as an increase in mean age per modal group of less than one day per daily time interval between samples. The data show that, in both females and males, the relationship between mean statolith increment number and elapsed time between samples is positive in some months and negative in others. The months in which the relationship between mean increment number and elapsed time is negative are similar to those months in which recruitment has previously been demonstrated to occur, in February, April/May and September for both females and males. Cohorts of squid remain available to the fishery from June–September (females) and February–May and June–July (males) thus allowing mean monthly growth rates of 0.4 mm d −1 for females and 0.7 mm d −1 for males to be calculated. The present study shows that putative age data derived from statoliths apparently demonstrate some sampling biases within a fishery and their effects on the measurement of growth. Migration through the sampled population is manifested by an increase in mean increment number of less than one per day between samples, and growth can only be assessed where the increase in mean increment number is not significantly different from the time elapsed between samples.