Physico-chemical, colorimetric, rheological parameters and chemometric discrimination of the origin of Mugil cephalus' roes during the manufacturing process of Bottarga

Abstract

The aim of this work was to measure the physico-chemical and the colorimetric parameters of ovaries from Mugil cephalus caught in the Tortolì lagoon (South-East coast of Sardinia) along the steps of the manufacturing process of Bottarga, together with the rheological parameters of the final product. A lowering of all CIELab coordinates (lightness, redness and yellowness) was observed during the manufacture process. All CIELab parameters were used to build a Linear Discriminant Analysis (LDA) predictive model able to determine in real time if the roes had been subdued to a freezing process, with a success in prediction of 100%. This model could be used to identify the origin of the roes, since only the imported ones are frozen. The major changes of all the studied parameters (p < 0.05) were noted in the drying step rather than in the salting step. After processing, Bottarga was characterized by a pH value of 5.46 (CV = 2.8) and a moisture content of 25% (CV = 8), whereas the typical per cent amounts of proteins, fat and NaCl, calculated as a percentage on the dried weight, were 56 (CV = 2), 34 (CV = 3) and 3.6 (CV = 17), respectively. The physical chemical changes of the roes during the manufacturing process were consistent for moisture, which decreased by 28%, whereas the protein and the fat contents on the dried weight got respectively lower of 3% and 2%. NaCl content increased by 3.1%. Principal Component Analyses (PCA) were also performed on all data to establish trends and relationships among all parameters. Hardness and consistency of Bottarga were negatively correlated with the moisture content (r = -0.87 and r = -0.88, respectively), while its adhesiveness was negatively correlated with the fat content (r = -0.68).

Keywords: CIELab coordinates; Linear discriminant analysis; Mullet roes; Penetrometry test; Physico-chemical characterization; Principal component analysis.