An objective of the AOTTP was to estimate Type-I (immediate) and Type-II (long-term) tagshedding rates for Atlantic yellowfin tuna from double-tagging experiments (4,518 double tags released with 1,061 recoveries). Accounting for the insertion point of the tag according to the body side of the fish, by introducing a tag-location effect in Type-1 (i.e., 1- α) and in Type-II tagshedding, in the constant-rate model did not improve significantly the fit. Type-I and Type-II tag shedding estimates (0.026 and 0.031, respectively) are close to the values obtained in the Indian Ocean (0.028 and 0.040, respectively). On the basis of these results, the shedding rate is about 6% the first year at sea and reaches 17% after 5 years at sea. Preliminary results suggested that tag loss could differ according to the size at release but additional factors must be analysed before drawing a definitive conclusion. This study showed that tag shedding rate should be taken into account with other sources of uncertainty such as the reporting rate in order to estimate exploitation and mortality rates derived from tagging data.