标题：Application of Principal Component Analysis to Characterize the Effect of Catching Ground on Post-mortem Quality Changes in Ice Stored Sea
Bream (Sparus aurata, L.) and Bogue (Boops boops, L.)
期刊名称：Croatian Journal of Food Technology, Biotechnology and Nutrition
印刷版ISSN：1847-3423
电子版ISSN：1847-7461
出版年度：2012
卷号：7
期号：SPECIAL ISSUE - 7th
页码：14-17
出版社：Hrvatsko društvo prehrambenih tehnologa, biotehnologa i nutricionista
摘要：Although limited, the existing literature indicates the significant difference in sensory and quality parameters of wild and farmed fish samples of the same species. Post-mortem quality changes of sea bream and bogue samples stored in ice were investigated by sensory (raw and cooked), physical (pH, dielectric properties) and chemical (lipid oxidation, volatile amines, proximate composition) analyses. Principal component analysis (PCA) was applied to study the relationships between the quality parameters during ost-mortem changes of farmed and wild sea bream and two “types” of bogues samples: samples aggregated around fi sh farm cages and natural control population. The PCA analyses showed distinct separation of wild and farmed/farm affected samples. The proportion of variance accounted for by the fi rst two principal components was 90.63 in sea bream and 89.35 % in bogue samples. Variations in storage time, sensory and chemical assessment were described by PC1, whereas variations in catching ground and proximate composition by PC2 in all investigated samples. Sensory evaluation, lipid oxidation index and volatile amine content of bogue samples showed high correlation with storage time; fat and moisture content, together with physical properties underwent the influence of the catching ground. Similar results were determined for sea bream, with exception for physical properties that showed higher correlation with the storage time in ice. Highest correlation with PC1 (>50 % of overall variance) were observed for sensory assessment (r>0.92) in both fish groups. Spoilage evolution over time given as the sum of all demerit points (quality index) showed higher correlation then any single parameters itself, indicating that the individual chemical and physical parameters could not replace the usage of sensory QIM scoring scheme in freshness assessment. PCA analyses were found useful in distinguishing the impact of fish farms on the wild fish populations.
