User menu
In 2018 the advice of yellowfin tuna in the Indian Ocean (YFT) was based on a grid of 24 models, where all models were based on the age and length structured integrated assessment model Stock Synthesis (SS). However, due to several issues in the data inputs and model assumptions, the Science Committee of IOTC (SC) recommended a workplan to improve the YFT assessment. Therefore, in this document, based on the comments of the WPTT21, two different processes were conducted: i) some of the basic assumptions on the assessment model were analyzed in details and ii) a new procedure on how to select the models to be included in the final grid used for the advice is presented.
The current model treats seasons as continuous years, and this complicates the settings of the model as well as the interpretation of the results. Therefore, with the aim of simplifying the model but at the same time improve the understanding of the modeling part of the key processes in the dynamic of the stock such as movement and recruitment, we transform the non-seasonal model into an annual model with seasons. The models were compared using diagnostics where the fits to the data, the prediction skills and the retrospective pattern are used to evaluate the performance of each model. The results are promising but still more works need to be done with the annual model before using it for assessment.
The other process analyzed in this study is the selection of the models to be included in the final grid used for advice. In WPTT21 the group discussed which were the main axis of uncertainties in the model assumptions and proposed a grid of models that could cover that uncertainty. In this study, based on that original grid we present different hypothesis that encapsulate the main axis of uncertainties of the assessment models, and present a new procedure leading to the selection of the models to be included the final grid used for providing the advice.
