Assessment of flesh colour in Atlantic salmon, Salmo salar L.

Abstract

The degree of pigmentation in muscle of Atlantic salmon, Salmo salar L., fillets of fish that were fed eight diets fortified with 10, 20, 40, 60, 80.100, 150 and 200 mg astaxanthin kg−1 and a non-supplemented control diet from 3 to 21 months was assessed using different methods. A tristimulus colorimeter (Minolta Chroma Meter) was used to measure the colour composition of the fillets instrumentally. The colour was also determined using the Roche Colour Card for Salmonids. The concentration of astaxanthin in the muscle was measured by chemical analyses. All measurements for colour were done directly on the epaxial muscle anterior to the dorsal fin. The lightness factor (L *). the red/green chromaticity (a*), the yellow/blue chromaticity (b*) and the saturation C* of the colorimetric readings and the Colour Card scores were compared with the chemical analyses. The astaxanthin concentration in the flesh varied from 1 to 10 mg kg−1 and the visual appearance of the fillets varied from yellowish-white to red. The relationship between the a*, b* and C* values and the astaxanthin concentration in the muscle was non-linear. Non-linear regression lines were found between the a* value and the astaxanthin concentration in the flesh (r2= 0.974) and the b* value and the astaxanthin concentration in the flesh (r2= 0.984). The instrument was not able to detect differences in astaxanthin concentration at astaxanthin levels above 3-4 mg kg−1 using the presented method directly on the fillet. The instrument might be useful for rejecting groups of salmon with poor pigmentation. A good linear regression was found between the Colour Card score and the mean astaxanthin concentration in the flesh (r2 - 0.992). The Colour Card provided a better prediction of the astaxanthin concentration at higher astaxanthin levels than the Chroma Meter. None of the methods provided a satisfactory prediction of the astaxanthin concentration in the muscle of individual fish using the presented methods.