Latin american journal of aquatic research
versão On-line ISSN 0718-560X
CANALES, Cristian e ARANA, Patricio M. Estimating yellow squat lobster (Cervimunida johni) biomass by applying a generalized linear model to catch records per swept area in central Chile. Lat. Am. J. Aquat. Res. [online]. 2012, vol.40, n.2, pp. 316-334. ISSN 0718-560X.
Changes in the annual density of the yellow squat lobster (Cervimunida johni) were modeled using a generalized linear model (GLM), using information obtained during direct stock assessment cruises carried out in central Chile (32°-38°S) between 1997 and 2009. The purpose of this approach was to minimize the effect of different sampling criteria and designs that have been used over time and to determine the factors explain in variations in the density of this resource. The above approach allowed estimating the expected value of the annual density for the biomass calculation through the "swept area" method. By using the haul-to-haul records of catch per unit of the swept area (CPUA), it was possible to identify and measure the points of abundance and characterize the distribution and area occupied by the resource over time. The significant effects identified by the GLM were the year, zone, depth, and first-order annual interactions; which explained 43% of the residual deviance in the model. The main effects that explain the CPUA variations are the year and the year-zone interaction. The zone with the greatest average abundance was located between 32° and 34°S; the same was for the depth range <250 m. Within the period analyzed, the biomass of the population of this resource increased steadily until 2006, mainly in the northern zone of the study area, followed by a sharp reduction 2009, that is 75% compared to the maximum and results in a biomass of 11,000 ton.
Palavras-chave : direct stock assessment; swept area method; CPUA; generalized lineal model; squat lobster; Cervimunida johni; Chile.