User menu
This paper describes work to estimate the age and growth of yellowfin tuna (Thunnus
albacares) in the Indian Ocean from otoliths as part of the ‘GERUNDIO’ project1. The 2018
stock assessment for yellowfin tuna in the Indian Ocean (IOTC) indicated that the stock isoverfished and subject to overfishing (Fu et al. 2018; IOTC 2020). The stock assessment model
used a fixed growth function from Fonteneau (2008) in the base model and additional growth
curves from Eveson et al. (2015) and Dortel et al. (2015) in sensitivity models. All of these
growth models suggest growth of yellowfin tuna is slow between 30 and 60 cm fork length
(FL) before changing to much faster growth between 60 and ~120 cm FL. The Fonteneau
(2008) model is based on growth information from tag-recapture data collected during the
Indian Ocean Tuna Tagging Programme (IOTTP), the Eveson et al. (2015) models integrate the
IOTTP tag-recapture data with otolith-based daily age estimates from Sardenne et al. (2015),
and the Dortel et al. (2015) models use different combinations of the tag-recapture and
otolith daily age data as well as length-frequency data from the purse seine catches. The
otolith-based daily age estimates from Sardenne et al. (2015) varied considerably among
readers, and there was a recommendation by Sardenne et al. (2015) to explore alternate
ageing methods such as annual ageing of otoliths (as opposed to daily ageing). Recently,
Farley et al. (2017; 2020) developed a new method to estimate the decimal age of bigeye tuna
in the western and central Pacific Ocean from validated counts of daily and annual growth
zones in otoliths. The aim of the current study is to apply this method to yellowfin tuna in the
Indian Ocean to obtain new estimates of age and growth, and to attempt to validate the age
estimates using otoliths and data from yellowfin tuna tagged and recaptured in the IOTTP.
Otoliths from 1479 yellowfin tuna collected in the current and previous projects were
available for analysis, ranging in size from 20.5 to 179 cm FL. Of these, 253 otoliths were
selected for ageing. A combination of daily and annual ageing was undertaken, and a final age
was obtained for 250 of the 253 fish. The youngest fish was aged 53 days and the oldest was
10.9 years. The preliminary age validation work using otoliths and data from the IOTTP
provides evidence that the otolith ageing method used in this study is accurate. However, we
recommend that further age validation work is undertaken, including the analysis of bomb
radiocarbon (14C) data from otoliths (currently underway as part of the GERUNDIO project)
and analysis of the OTC marked otoliths by a reader with no prior knowledge of the time at
liberty or fish length.
Four growth models were fit to the age and length data (von Bertalanffy (VB), Richards, VB
log k, and 2-stage VB). All four models provided very similar fits; however, the 2-stage VB
model provided a better fit to the data for fish ~<55 cm FL. The length-at-otolith weight data
(which is independent of the age estimation method) showed a change in otolith growth at
~55 cm FL, which is consistent with the length-at-age data and lends support to the 2-stage
VB model. Overall, our analysis shows that growth is rapid in the first few years with fish
reaching ~60 cm FL at age 1 and ~95 cm FL at age 2. Mean asymptotic length was estimated
to be ~163 cm FL.
The 2-stage VB growth curve estimated in the current study is similar to growth estimated by
Multifan-CL (MFCL) in the 2008 stock assessment for fish ~<90 cm FL. The divergence in
growth for fish larger than 90 cm FL is not surprising since MFCL relies on length frequency
data from catches to estimate growth parameters, which are imprecise when length modes
merge across age classes as fish grows. In contrast, our two-stage VB growth curve is quite
different from the “ad hoc” growth curve of Fonteneau (2008) and VB log k growth curves of
Eveson et al. (2015) and Dortel et al. (2015). In particular, we do not see the same slow growth
for fish <60 cm, followed by a rapid increase. Furthermore, we estimate mean asymptotic
length to be much higher (~163 cm FL) compared to the growth curves that included tagrecapture
data (130.7 cm FL in Eveson et al. 2015; ~139 cm FL in Dortel et al. 2015). This may,
at least in part, be due to the low number of fish >150 cm FL in the tag-recapture data
available at the time to be used in Eveson et al. (2015) and Dortel et al. (2015), which is likely
related to the relatively short times at liberty of fish included in the analysis (<6 years)
compared to the current estimated longevity of yellowfin tuna of at least 10 years.
We recommend that additional otoliths are collected from the northern and eastern regions
of the Indian Ocean, and that these otoliths and additional otoliths from those already
collected in the GERUNDIO and IOTTP projects are read/aged to provide further information
on growth and longevity. These data will also be useful for assessing the potential
