The pink colour of the flesh of Atlantic salmon is highly favoured by consumers and is therefore an important quality parameter. However, there is considerable evidence that undesirable colour changes occur in Atlantic salmon fillets during processing into a cold-smoked product and during subsequent storage. As astaxanthin is considered to be responsible for the pink colour of Atlantic salmon flesh, this colour change may be expected to be due to the oxidative degradation of this carotenoid. This project was designed to examine this colour change and changes to astaxanthin and other components of the fish flesh which may be involved in such oxidation reactions. The other components included the anti-oxidative vitamins α -tocopherol and ascorbic acid, moisture, lipid (total and individual fatty acids) as well as lipid oxidation products as measured by thiobarbituric acid-reactive substances (TEARS). The inter-relationships among these components and between these components and measures of colour were examined in order to identify the causes of the colour change during processing and storage.
The changes in the fillets were investigated in 3 experiments: in Experiments 1 and 2, both fresh and previously frozen fish were used whereas in Experiment 3 only fresh fish were used. For Experiments 1 and 2, conducted in August 2002, samples of Tasmanian Atlantic salmon (Salmo salar) stored frozen since harvest in February 2002 (designated as "frozen") together with fresh fish harvested at that time (designated as "fresh") from Tasmanian salmon farms were used. Fillets from both frozen and fresh fish were used in order to ascertain whether prior freezing affected the extent of colour and other changes in the flesh during processing and subsequent storage. For Experiment 3, fresh fish harvested in January 2004 were used without being frozen.
In Experiment 1, the colour attributes of salmon flesh were examined during standard commercial processing practices from 'raw' (A) to 'washed' (B) to 'salted' (C) to cold 'smoked' (D) fillets. Flesh colour at each step was recorded objectively using a Minolta Chromatometer and the colour of raw fillets was measured subjectively using the SalmoFan colour card. The colour card scores of raw fillets from the frozen and fresh fish were not significantly different (27.1 and 27.3 for frozen and fresh fish respectively), while the measured redness (a*) values were significantly different, around 13 and 10 for raw frozen and fresh fillets, respectively. During processing, the instrumental colour score, especially a* value (redness), decreased significantly. The content of astaxanthin, α-tocopherol, lipid and individual omega-3 fatty acids remained constant during processing, but ascorbic acid and moisture contents decreased significantly while TEARS values increased significantly.
Significant positive correlations were observed between redness (a* value) and astaxanthin, and between lipid and α -tocopherol before smoking, but not after smoking, while significant relationships between α -tocopherol and astaxanthin, and α -tocopherol and ascorbic acid were only observed for raw fillets. There was no clear relationship between lipid and astaxanthin contents.
In Experiment 2, the colour attributes were investigated during retail display of vacuum-packaged cold-smoked salmon (E = retail display day 0, F = retail display for 14 days and G = retail display for 28 days). The only significant changes found under the simulated retail display conditions used were instrumental colour score, especially redness (a* value) and ascorbic acid of frozen fish after 28 days' display.
The higher astaxanthin concentration appeared to be associated with higher redness values (a*) but in almost all cases, the correlation did not reach statistical significance. Overall, α -tocopherol and ascorbic acid showed a significant relationship but no clear correlation was found between α -tocopherol and astaxanthin, lipid and astaxanthin, α -tocopherol and lipid, and lipid and TEARS value during retail display for 28 days.
In Experiment 3, colour attributes during processing and subsequent chilled storage of the unpacked, cold-smoked salmon fillets prior to marketing were investigated. This experiment was designed to simulate commercial practice where cold-smoked fish are stored chilled for 10 to 14 days prior to being sliced and vacuum-packed for retail sale. From the results of Experiments 1 and 2, it appeared that changes during the period of chilled storage between processing and retail display were significant and warranted investigation. Considerable oxidation occurred in cold-smoked salmon between the end of Experiment 1 (sample D) and day 0 of the retail display in Experiment 2 (sample E); this led to Experiment 3 being conducted. Processing practices in the factories changed while this project was being conducted, so that in Experiment 3 one of the processing steps examined in Experiment 1, washing (B) was not applied. Samples examined were: 'raw' (A), 'salted' (C), cold-'smoked' (D), cold-smoked stored chilled for 3 days (Dl), cold-smoked stored chilled for 7 days (D2) and cold-smoked stored chilled for 10 days (E), equivalent to retail display day 0.
In Experiment 3, the instrumental colour score, especially redness, decreased significantly during processing, confirming the discolouration found in Experiment 1. The lack of a washing step in Experiment 3 did not appear to affect the concentrations of astaxanthin, a-tocopherol, lipid and TEARS in smoked salmon (D), but did appear to affect ascorbic acid and moisture contents. The only significant interactions found in this experiment were between α -tocopherol and astaxanthin, and between α- tocopherol and lipid immediately after smoking (D).
The results indicate that cold smoking of Atlantic salmon fillets has a significant effect on surface colour, regardless of the application of a washing process step. Astaxanthin concentrations were stable throughout processing, subsequent chilled storage prior to marketing of cold-smoked salmon and during retail display. Alphatocopherol, total lipid and percentage of individual fatty acids, including EPA and DHA appeared to remain constant in the experiments conducted. Processing followed by chilled storage in unpackaged form significantly increased the oxidation status (TEARS value), and decreased ascorbic acid concentration and moisture content. However, the TEARS level, moisture and ascorbic acid contents generally remained constant during retail display for 28 days. In most cases, there was similarity in the patterns of component changes of frozen and fresh fillets, suggesting that the effect of processing is greater than the effect of freezing.
It is concluded that the change of colour during the standard commercial processing steps and following chilled storage resulted from changes on the surface of the flesh only. The changes in the instrumental colour measurements, especially redness (a* value), could not be explained by changes in the pigment astaxanthin. Despite the surface colour change and some oxidation occurring in the salmon before retail sale, the levels of astaxanthin and omega-3 fatty acids were not affected and the product was still of high quality. The changes observed may be able to be minimised by applying a vacuum packaging step immediately after processing.