Pelagic longline fishing operation parameters optimization - A case study on targeting yellowfin tuna (Thunnus albacares) in the Indian Ocean

In longline fishery, in order to improve fishing efficiency, it was necessary to accurately control the depth of hooks to set the hook as far as possible in the preferred water layer of target species. In this paper, the catenary hook depth formula was used to calculate the theoretical depth of the hook. The environmental data, e.g. wind speed (V_w), gear drift velocity (V_g), angle of attack (Q_w) (the angle between the prevailing course in deploying the gear and direction that the fishing gear was drifting), the wind angle (γ) (the angle between the direction of the wind and the prevailing course in deploying the gear), and operation parameters, e.g. line shooting speed (V₁), vessel speed (V₂), the number of hooks between two floats (N_b), and time interval between two hooks (t), were collected and the actual hook depth (D_f) were measured on board of the longliners. The multiple linear stepwise regression analysis was used to establish the hook depth prediction model with the theoretical depth , the environmental data and operation parameters as the independent variables and the measured hook depth (D_f) as the dependent variable to predict hook depth (D_f'). This paper used Matlab software to code the operation parameter calculation, calculate the predicted hook depth (D_f) within the preset range of each parameter, respectively and select the range of operation parameters when the hook numbers distributed in the preferred water layer (100m-160m) of yellowfin tuna was maximum. The results show that: 1) the prediction model of hook depth was , where,  was the ordinal number of hook counted from one float side and other parameters as above; 2) the number of hooks distributed in the water layer of 100m-160m increased with the increase of N_b. When N_b =17, the percentage was the smallest (76.5%), and when N_b =27, the percentage was the largest (85.2%); 3) when the number of hooks between the two floats was 17-27, the hook depth mostly concentrated in the depth range of 140-160m; 4) when the number of hooks was the highest in the water layer of 100-160m, the operating parameters, e.g shortening rate (k), line shooting speed (V₁), the vessel speed (V₂) and the wind speed (V_w) decrease with the increase of the number of hooks between the two floats (N_b). This paper suggested that: 1) the multiple linear stepwise regression analysis can be used to accurately predict the hook depth; 2) results suggested that the N_b should be 20-23, k should be 0.77-0.9, V₁ should be 5-7m/s, V₂ should be 3.8-5 m/s, t should be 6-8s, V_w should be 0-6.5m/s, to ensure the percentage of hooks distributed in the range of 100-160m higher than 80%; 3) optimizing the depth of hooks according to this study would help to reduce bycatch of some sharks and turtles, etc.