Crescimento polifásico de peixes : relações ecológicas e taxonômicas

• Autor(a) principal: Barradas, José Ricardo de Souza
• Data de Publicação: 2016
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da PUC_RS
• Texto Completo: http://tede2.pucrs.br/tede2/handle/tede/6791
• topic: PEIXES - ESPÉCIES; BIOLOGIA; ZOOLOGIA; CIENCIAS BIOLOGICAS::ZOOLOGIA

Length-weight data are the basis of most fish sampling programs and an important tool for fishing biology. Contradictorily, very often research on the relationship between weight and length in fish are disclosed only as brief publications on species for which this information is not yet known. The largest contribution to weight modeling and length was made by Huxley, when a power equation was used to describe the Allometric growth. Although Huxley model has been systematically used in animal growth studies since then, this proposal has limitations in estimating the allometric factor (“b”) as constant throughout the life cycle. Several authors have identified variations in allometric coefficient for animal development, being these complex patterns resulting from various factors and eventually obscured due to natural variability of the data. The objective of this study was to identify growth patterns in fish by using a model that considers changes in the growth occurring during the life cycle and through the crossing of different information of the species biology. We sought to understand the relationships between the polyphasic growth and the achievement of sexual maturity, in the form of migration, with habitat use patterns and the families which belong to the species of study. The data used were obtained from two fronts: (1) review of the database already available in Aquatic Ecology Laboratory of the Catholic University of Rio Grande do Sul (PUCRS) and (2) contact with researchers from other institutions. As a final product, there was obtained a database composed of 78 species divided into 35 families, comprising an array of 92,889 points. The data were imported in the statistical R platform, where an automated algorithm was developed and implemented to adjust the equations. A total of 109 estimates of weight-length were obtained. In general, the average allometric coefficient were slightly larger than 3, for the uniphasic model and for both phases of the polyphasic model, indicating a positive allometry. Multi-modality in the frequency distribution of ! was observed in Huxley's model, this behavior wasn't identified in both phases of polyphasic model. The final adjustment obtained by polyphasic model was satisfactory when evaluated in the context of a large number of species. Strong correlation has been identified between the SCP and (1) the maximum sampled sizes (r2 = 0.94), (2) estimations of first maturity (r2 = 0.93) and (3) length at first maturity (r2 = 0.96). This behavior was also identified between the length at first maturity and estimates of first maturity (r2 = 0.97). The Stanza Changing Points (SCP) were overestimated by approximately 15% when compared with the estimation of first maturity. Estimation of first maturity were overestimated by approximately 5% when crossed with the lengths at first maturity. Patterns were identified in the distribution of homocedastic residuals for species of lotic (p < 0.001) and lentic (p = 0.023) habitats and a significant difference between lotic and neritic environments (p = 0.036). Regarding the form of migration, a pattern was identified in the distribution of error for potamodromous species (p = 0.001) and significant difference between the oceanodromous and potamodromous species (p = 0.023). It wasn’t found any kind of pattern or trend between the SCP and the estimation of first maturity when separated by families. The application of statistical and computational tools developed in this work makes it possible to establish important relationships to improve the understanding of fish growth. Given the ease of obtaining length-weight data, explore this information mathematically can increase understanding of populations of economic and ecological interest, allowing for constantly updating and maintenance information for environmental management and fishing, in addition to stock control.