Avaliação de características de desempenho e morfológicas do camarão Penaeus vannamei por meio de análise de componentes principais

Abstract

The cultivation of the marine shrimp Penaeus vannamei represents a sector of great economic and social relevance, requiring constant optimization of production systems. The evaluation of zootechnical performance is a complex process that involves the measurement of multiple correlated biometric characteristics, hindering the identification of practical indicators and efficient decision-making. This doctoral thesis aimed to evaluate the performance and morphometric variables of Penaeus vannamei at different ages and strains, cultivated in excavated ponds, using Principal Component Analysis (PCA) as a statistical tool to synthesize information and identify the main factors of variation. The research was structured in two articles, in which PCA proved to be a robust and effective methodology. In both studies, the technique was able to reduce the dimensionality of the data, concentrating most of the total variance in a single overall performance index, the First Principal Component (PC1), which consistently explained more than 82% of the observed variation. This result shows that shrimp performance is a predominantly one-dimensional phenomenon, simplifying the approach to management and genetic improvement. Article 1 focused on evaluating performance at different cultivation ages. The results indicated that Total Length (TL) is the biometric indicator with the greatest contribution to PC1, and is, therefore, the most efficient for routine monitoring on farms. Furthermore, Fillet Weight (FW) was identified as the most relevant performance characteristic, reflecting its importance as a quality attribute of the final product and a comprehensive indicator of growth. The analysis also revealed that the first two principal components (PC1 and PC2) accounted for 95.91% of the total variance. Article 2 addressed the evaluation of performance in different strains and harvest ages. PCA synthesized the nine biometric variables, with Live Weight (LW) being the most representative variable of the overall performance index (PC1, explaining 84.15% of the variance). The most impactful finding in practical terms was the observation that zootechnical performance at 45 days of cultivation is comparable to performance at 60 days. This discovery suggests the possibility of anticipating the harvest, which constitutes a more economical and efficient management strategy, capable of reducing operational costs, decreasing the time of exposure to sanitary risks, and increasing the annual productivity of shrimp farming. In conclusion, this thesis validates PCA as a powerful tool for performance discrimination and decision-making in aquaculture. More significantly, the thesis offers a scientific basis for the anticipation of harvest to 45 days, promoting greater sustainability and profitability for the sector. The information generated is fundamental for the development of more efficient genetic improvement programs and for the consolidation of more agile and scientifically based management.