INTRODUCTION

The global tuna fishery is of great economic importance. According to ^{Xolaltenco-Coyotl et al. (2010)}, most commercial catches of tuna species are made in the Pacific Ocean (69.0% of the total catch in 2007). The main species caught in the eastern Pacific Ocean is the yellowfin tuna (YFT), making up 90.0% of the total annual catches made in that region (^{DOF 2015}).

In Mexico, tuna are harvested with purse seines. However, a longline YFT fishery was developed in recent years. In 2012 YFT exploitation with longlines began as bycatch fishing in the shark fisheries operating on the coast of Nayarit, and in time the longline fishery targeting YFT arose. According to information from cooperatives, in 2013 authorities granted commercial permits issued by the Mexican National Commission for Aquaculture and Fisheries (CONAPESCA, for its acronym in Spanish). Because the longline tuna fishery in Nayarit is relatively new, there is no regulation for this specific fishery.

The tuna longline fishery based in Banderas Bay, Nayarit, operates throughout the year, with an annual production of 66.30 t. About 30 small ships operate regularly, but up to 80 small ships can operate occasionally. The fleet includes 9.14-m (30.00-ft) long BOOGIE type boats with two 150-hp outboard motors. The fishery uses an artisanal longline.

This paper presents initial results on basic biological information regarding length-weight relationships, size and weight structures, and selectivity. This information contributes to developing regulations for the small-scale tuna fishery operating with longlines in marine waters off the coast of Nayarit in order to build the basis for the sustainable exploitation of YFT.

MATERIALS AND METHODS

The fishing area is located along the coast of Nayarit and near the Isabel Island and the Marias Islands in Mexico. YFT were caught during July-August and November-December 2015 and January-March 2016. Specimens were measured (cm) and weighed (kg).

We fit the length-weight relationship using the following power function:

where W is the weight, L is the total length, a and b are parameters to be estimated. We fit the model (Equation 1) using the nls function in the R statistical package (^{R Core Team 2020}). We calculated the 95% confidence interval for the b parameter to determine the type of growth with the one-way t-test.

To estimate selectivity we used the logistic selectivity curve (^{Millar and Fryer 1999}):

where S(L) is the retention probability for YFT at L, a and b are shape parameters, so that the 50% retention probability is reached at length −a/b, and δ is an asymmetry parameter. We estimated the parameters and their uncertainties using the negative log-likelihood (LL) and assuming a normally distributed observation error:

where n is the number of observed data, S _i ^obs represents the observed selectivity for a given length, S _i ^est represents the predicted selectivity, and σ is the standard error. The estimation was carried out using the nls function in the R statistical package (^{R Core Team 2020}).

RESULTS

We measured 584 YFT, representing a catch of 25.5 t. The total length l values varied from 73.0 to 228.0 cm, and the mean (±SD) was 153.0 ± 23.5 cm (Fig. 1a). The W values ranged from 5.8 to 128.0 kg, and the mean was 43.7 ± 19.4 kg (Fig. 1b).

Figure 1 Distribution of yellowfin tuna total length (a) and weight (b); broken line represents length at maturity and weight at maturity, respectively. 

The length-weight relationship (Equation 1) fit the data well. The 95% confidence interval for b was (2.88, 2.96). A t-test (H ₀: b = 3) suggested that YFT could exhibit negative allometric growth. The 95% confidence interval for a was (1.43 × 10^-5, 2.14 × 10^-5) (Fig. 2a, Table 1).

Figure 2 Length-weight relationship fit (a) and selectivity at length (b) for yellowfin tuna in Nayarit, Mexico. Obs, observed data; Pred, predicted data; and CI, 95% confidence interval. 

The asymptotic selectivity model fit the observed data well (Fig. 2b). The shape parameters were a = -7.660 ± 1.370 (standard error) and b = 0.057 ± 0.005, and the asymmetry parameter was δ = 0.488 ± 0.230. The correlation coefficient between the observed and the estimated selectivity was 0.950 (P value ~ 9.569 × 10^-8).

DISCUSSION

The YFT exploitation started in 2013, so it presents a rare opportunity to establish a management plan for the fishery in its predevelopment phase. Some key calculations depend on data gathered during this phase, such as potential yields, age structure, size structure, and the natural mortality, which can be measured only in this phase (^{Hilborn and Walters 1992}). The tuna longline fishery status can be assessed when compared with the purse seine fishery. The purse-seine catches fluctuated between 10,000 and 25,000 t per year from 1988 to 1997 (^{Trigueros-Salmerón 2003}). The longline tuna average landing was 66.3 t (2015-2017), representing a small percentage of the purse seine catch.

Our results suggest that the size structure in the catch corresponds to an adult population, since less than 1.0% of tuna measurements were below the length at first maturity of 95.0 cm (^{Suzuki 1994}). Therefore, this artisanal fishery does not contribute to growth overfishing. The largest organisms arrive at the study area during July and August, suggesting that fishing should be concentrated in the summer months. However, maturity studies must also be considered when developing fishing closure regulations to protect spawning. This result agrees with López-Medina’s (2004) suggestion that free schools of adult fish are located near the coast. It is worth noting that the tuna in our survey were larger than the tuna caught by the longline fishery in the eastern Pacific (^{Ortega-García 1996}), where the largest YFT measured 185.0 cm, whereas the largest YFT caught by us in Nayarit measured 228.0 cm; 8.2% of our tuna measured over 185.0 cm in length.

The length-weight relationship analysis for YFT showed a negative allometric growth pattern (b = 2.91 ± 0.02), where growth in length is slower than growth in weight and species become slender as length increases (^{Pauly 1984}). Our result agrees with estimates for the same species in the China Sea (^{Ma et al. 2016}) and the Indian Ocean (^{Rohit et al. 2008}). This information is key for future management strategies because length-weight relationships are commonly used in stock assessment, for example, for converting weight frequencies into lengths and raising length samples to length frequencies for catch-at-length tables (^{Ward and Ramirez 1992}). The YFT length-weight relationship can change depending on the area, year, and sex (^{De Giosa et al. 2014}); therefore, we need to keep collecting data to use it in future fishery assessment and management.

The selectivity fit suggests that selectivity follows an asymptotic pattern, indicating that the fishery catches all fish greater than a certain size in proportion to their occurrence in the population (^{Piner 2012}). Our results are in agreement with the ones found for the Taiwanese and Japanese longline fleet catching pacific bluefin tuna (^{Piner 2012}).

The results presented here represent the first step into gathering information to establish a research program and a monitoring program for YFT on the Pacific coast of Mexico. Nevertheless, it is necessary to establish a new monitoring program that permanently collects information on population parameters such as longevity, sexual maturity, relative abundance indices, growth, feeding habits, bycatch, and ecosystem considerations. This information is vital for reliable stock assessments and management plans, and the sustainable development of fisheries (^{Hoggarth et al. 2005}). It may be useful in managing the rapidly developing YFT fishery. Future research programs must focus on stock assessment and ecosystem approach to fisheries management. The establishment of these programs will help develop the necessary fishing regulations to ensure a well-managed fishery, a potential certification, and the sustainable exploitation of the YFT.