A new approach to modelling the shelf life of Gilthead seabream (Sparus aurata)

A total of 217 Gilthead seabreams were subdivided in four groups, according to four different storage conditions. All fish were evaluated by both Quality Index Method (QIM) and microbiological analysis, sampling skin, gills and flesh, separately. A QIM score predictive system was set by modelling the growth of microflora of skin, gills and flesh and coupling these predictions to each related partial QIM score (QIMSkin, QIMGills, QIMFlesh). The expression of QIM score as a function of bacterial behaviour was carried out by the employment of two coefficients. The predicted mean bacterial concentrations corresponding to the QIM score at 14 days were always near to Log 8 CFU g^{ -1} in the case of 'S' (skin) and 'G' (gills) series. Moreover, predicted QIM scores were in a good agreement with observed data, reproducing the observed mean time of rejection as well as the bacterial spoilage level (Log 8 CFU g^{ -1}), for all kinds of storage condition.