Determining the age of tropical tunas in the Indian Ocean from otolith microstructures

• We confirm that yellowfin and juvenile bigeye deposit daily micro-increments in otoliths.
• Our results support previous findings that deposition rates in adult skipjack are highly variable.
• Age estimates are characterized by very low precision and strongly dependent on the reader.
• Variability in otolith reading can affect growth estimates.
• Reference collections are required to standardize the reading of tuna micro-increments.

The Indian Ocean Tuna Tagging Program (IOTTP) provided a unique opportunity to assess the viability of estimating the age of tropical tunas from the micro-structural features of otoliths. Here, we analyzed the length measurements and micro-increment counts collected for 506 sagittal otoliths, of which 343 were chemically marked with oxytetracycline, for bigeye (Thunnus obesus), skipjack (Katsuwonus pelamis), and yellowfin tuna (Thunnus albacares). Our results show that the otoliths of tropical tunas grow more slowly than the rest of the body. Our findings confirm that both yellowfin and juvenile bigeye deposit daily increments in their otoliths, though ages are underestimated for large bigeye (>100 cm) when derived from micro-increment counts. Our results also indicate that skipjack otoliths are not suitable for age estimations during the adult phase, as evidenced by the poor agreement between micro-increment counts and days-at-liberty. We hypothesize that the income breeding strategy of skipjack could explain the variability observed in the deposition rates. Due to their complex micro-structural patterns, the reading of tropical tuna otoliths requires a degree of interpretation that can result in poor count precision and large variability in micro-increment counts, both among and within teams of readers. Age estimates were found to vary between readers, a factor which can eventually affect growth estimates and ultimately, impact on fisheries management decisions and outcomes. To address this, we recommend that reference collections of otoliths are developed, with a view to standardizing the reading process. Further, alternative methods, such as annual age estimations (as opposed to daily), and alternative structures, such as dorsal spines for skipjack, should be used to improve the accuracy of age estimations and the speed with which they can be made.