Age and growth of skipjack tuna (Katsuwonus pelamis) in the western equatorial Atlantic based on dorsal spines analysis / Idade e crescimento do bonito-listrado (Katsuwonus pelamis) no Atlântico Oeste Equatorial com base na análise dos espinhos dorsais

The skipjack tuna (Katsuwonus pelamis) is a migratory pelagic fish occurring in all tropical and subtropical oceans of the world. Due to its economic importance and the unbridled increase in fishing efforts, stocks may collapse if this resource is not managed properly. The present study aimed to estimate growth parameters from different growth models based on annual rings from the dorsal spines of skipjack tuna caught in the western equatorial Atlantic. The first dorsal spine of each individual was extracted to produce cross-sections with 0.6 to 0.8 mm in thickness. We used a multi-model approach to select the best growth model based on the Akaike information criterion (AIC). Two hundred seventy individuals were analyzed, with an average length of 49.58 ± 7.72 cm. The von Bertalanffy growth model had the best fit to the data, but the Gompertz and Logistic models also had essential support. Average asymptotic length (LL) was estimated to be 114.05 and 102.63 cm for observed and back-calculated data, respectively. The dorsal spines are indeed efficient tools for estimating growth parameters and multi-model inference is a novel approach for adjusting discrepancies that likely result from a single-model approach.

O bonito-listrado (Katsuwonus pelamis) é um peixe pelágico migratório que ocorre em todos os oceanos tropicais e subtropicais do mundo. Devido à sua importância econômica e ao aumento desen-freado do esforço de pesca, os estoques podem entrar em colapso se este recurso não for manejado adequadamente. O presente estudo teve como objetivo estimar parâmetros de crescimento para di-ferentes modelos baseados nos anéis etários formados nos espinhos dorsais do bonito-listrado cap-turado no Atlântico Oeste Equatorial. O primeiro espinho dorsal de cada indivíduo foi extraído para produzir cortes transversais com 0,6 a 0,8 mm de espessura. Utilizamos uma abordagem de múltiplos modelos para selecionar o melhor modelo de crescimento com base no critério de informação de Akaike (AIC). Foram analisados 277 indivíduos, com comprimento médio de 49,58 ± 7,72 cm. O modelo de crescimento de von Bertalanffy teve o melhor ajuste aos dados, mas os modelos de Gompertz e Logístico também tiveram um suporte essencial. O comprimento assintótico médio (LL) foi estimado em 114,05 e 102,63 cm para dados observados e retrocalculados, respectivamente. Os espinhos dorsais são de fato ferramentas eficientes para estimativa dos parâmetros de crescimento e a inferência de multi-modelos é uma nova abordagem para ajustar as discrepâncias que provavelmente resultam de uma abordagem com base em um único modelo.

Modelización de la captura y fases de desarrollo de la pesquería de la jaiba azul (Callinectes sapidus) en la Laguna Madre, Tamaulipas, México / Modelling the catch and development phases of the blue crab fishery (Callinectes sapidus) in the Laguna Madre, Tamaulipas, México

RESUMEN La captura promedio anual de la pesquería de la jaiba azul (Callinectes sapidus) (JA) en Tamaulipas, México se estima en 2 733 T, de la cual, el 82 % se pesca en la Laguna Madre, sitio que se considera aprovechado al máximo de su capacidad. El objetivo de la presente investigación fue modelar la captura anual de la JA en la Laguna Madre, Tamaulipas, mediante el ajuste de funciones matemáticas de tipo lineal y no lineal (o curvilínea), a la serie de tiempo de 1998 a 2012, además de identificar las fases de desarrollo de la pesquería, de acuerdo a varios modelos generalizados. Se utilizó el enfoque de la teoría de la información y el procedimiento de la inferencia multimodelo (IMM). Se ajustaron 11 modelos de regresión lineal y no lineal. Para la selección de modelos se utilizaron los criterios de información Akaike corregido (CIAc) y bayesiano (CIB). Para el IMM se consideró el nivel ∆i <2de plausibilidad de CIAc y CIB. Los modelos elegidos para el IMM fueron compuesto, crecimiento, exponencial, logístico, potencial y el sigmoideo, considerándose como más adecuados los primeros cuatro modelos citados. Los modelos promedio del IMM presentaron valores de β 0 y β 1 de 0.939 y 0.377 respectivamente, según CIAc; y de 0.952 y 0.344 respectivamente, de acuerdo al CIB. Solo los modelos compuesto y logístico mostraron significancia estadística en sus dos parámetros de regresión (β 0 y β 1). El índice de sustentabilidad pesquera reveló seis periodos de la captura y una disminución en magnitud de los cambios de la captura. La serie de datos analizada incluye dos ciclos de vida de acuerdo a los modelos de Csirke y Caddy. Los resultados mostraron que al final del periodo estudiado la pesquería se encontraba en colapso y decadencia.

ABSTRACT The average annual catch of the blue crab (Callinectes sapidus) (BC) fishery in Tamaulipas, Mexico is estimated at 2 733 T, of which the 82 % is caught in the Laguna Madre, which is considered to be exploited to the maximum of its capacity. The objective of the present investigation was to model the annual catch of the BC in the Laguna Madre, Tamaulipas, by adjusting mathematical functions of the linear and nonlinear (or curvilinear) type, to the time series from 1998 to 2012. In addition, the phases of development of the fishery, according to several generalized models, were identified. We used the information theory approach and multimodel inference procedure (MMI). Eleven linear and nonlinear regression models were fitted. For the selection of models, the corrected Akaike corrected (AICc) and Ba-yesian (BIC) information criteria were used. For the MMI, the level ∆i < 2 of plausibility of AICc and BIC was considered. The models chosen for the MMI were compound, growth, exponential, logistic, potential and the sigmoid, with the first four models being considered the most suitable of all those cited. The average models of the MMI presented values of β 0 and β 1: 0.939 and 0.377 respectively, according to CIAc; and 0.952 and 0.344 respectively, according to the CIB. Only the composite and logistic models showed statistical significance in their two regression parameters (β 0 and β 1). The fishery sustainability index revealed six catch periods and a decrease in magnitude of catch changes. The data series analyzed includes two life cycles according to the Csirke and Caddy models. The results showed that at the end of the studied period the fishery was in collapse and decay.