SCIENTIFIC COMMITTEE SEVENTH REGULAR SESSION 9-17 August 2011 Pohnpei, Federated States of Micronesia TAG REPORTING RATE PRIOR DISTRIBUTIONS FOR THE 2011 BIGEYE, YELLOWFIN, AND SKIPJACK STOCK ASSESSMENTS

Introduction A tag reporting rate is a critical parameter required for modelling tag recapture data and defines the probability that a tag is returned, given that a tagged fish was recaptured. The tag reporting rate directly scales the number of returned tags, and therefore directly affects the estimate of fishing mortality. In the absence of other information, knowing only total catch and the proportion of released tags that were returned, the biomass estimate is directly proportional to the tag reporting rate (RR), and the fishing mortality estimate is proportional to 1/RR. The tag reporting rates used in MFCL are affected by the following factors. 1. Tagging-related mortality and tag shedding before capture 2. Non-reporting of tags that have been captured This purpose of this paper is to provide estimates for input to the MULTIFAN-CL stock assessments for bigeye, yellowfin, and skipjack tuna. Methods Tagging data were analyzed for bigeye, yellowfin, and skipjack to estimate two major components of tag reporting rates: tag-related mortality and tag shedding before capture (i.e. at sea), and non-reporting of captured tags (i.e. on board or at the cannery). The first component was investigated in two stages. First, average return rates were estimated relative to a tag that was well-placed in a well-treated fish by the most experienced tagger, by comparing the effects on return rate of tagging characteristics such as the tagger identity, fish condition, and tag placement quality. Second, 'base' rates of tag mortality and tag shedding for this 'ideal' condition were assigned. Tag shedding was estimated from double tagging experiments, in which tags were well placed by the most experienced taggers, and assumptions were made about tag-related mortality rates. Captured-tag reporting rates for the PTTP were estimated via an analysis of data from tag seeding experiments, while for the RTTP they were available from previous work (Hampton 1997). Finally, the components were combined using a Monte-Carlo process to give an overall reporting rate prior distribution. Tag related mortality and loss Relative return rates The effects on tag return rate of tagger, release condition and tag placement were estimated for tags placed during the Relative return rates were modelled by tagging event using a binomial GLM and analyzed in the statistical package R (Ihaka and Gentleman 1996). Releases with the following characteristics were removed: fish length <=15cm, tag quality of 'tag lost', tag rejected', or 'unknown'. The following variables were included in the analysis: …