Sources of variation in the Trentingrana cheese production chain

Protected Designation of Origin (PDO) is a suitable tool to protect the reputation of regional foods and to eliminate the unfair competition and misleading of consumers by non-genuine products. Moreover, the achievement of the PDO label represents an opportunity to reach a premium price for the product, particularly in disadvantageous areas such as the Alps where the relationship between local economy and environment preservation is strong. More than 50% of the milk produced in Italy is processed into PDO cheeses and the most popular are Grana Padano and Parmigiano Reggiano. Trentingrana is a Grana-type PDO cheese manufactured in the mountain area of the Trento province. Although this label belongs to the most popular Grana Padano, cheese making protocol rules are more strict and similar to those of Parmigiano Reggiano. This thesis is based on the results of a wide research project aimed at monitoring the whole Trentingrana cheese production chain. Specific objectives of this work were a) to investigate sources of variation of herd bulk milk destined to Trentingrana cheese production and monthly farm quality score used in the milk payment system (Chapter 2), b) to study factors affecting the quality classification of Trentingrana wheels (Chapter 3), and c) to investigate sources of variation of sensory traits of ripened Trentingrana cheese and the quality index used for payment of the wheels (Chapter 4). The thesis starts with a general introduction on international and national dairy sector (Chapter 1), followed by three contributions (Chapters 2, 3 and 4), and general conclusions (Chapter 5). Chapter 2 assessed the impact of dairy factory, year of production, and month of production on milk yield and quality traits of milk used to produce Trentingrana cheese and the monthly farm quality score adopted to calculate the farm milk price. Relevant differences among dairy factories were found, particularly for milk yield. The year of production had a strong influence on milk yield and much less on quality aspects. Finally, the month of production exerted a significant influence on milk yield and exbibited a circannual variation, whereas the influence on quality traits was less pronounced. The milk farm quality score showed high variability and was significantly affected by all the effects included in the model. levant differences among dairy factories were detected. Chapter 3 focused on factors affecting the quality classification of Trentingrana cheese wheels. Traits considered were the percentage of first quality wheels on total wheels examined at 9 months of ripening, the percentage of first quality wheels on total wheels examined at 18 months of ripening and the percentage of first quality wheels at 18 months of ripening on the number of wheels evaluated at 9 months. The experimental unit was the batch of 2-months of production of each of 10 cooperative dairies from 2002 to 2008. An analysis of variance was carried out to investigate the influence of dairy factory, and year and season of production on the above traits. All factors significantly influenced cheese quality classification, with the dairy factory being the most important source of variation, followed by season and year of production. The first 4 years of production had a relevant and negative impact on the percentage of wheels chosen as first quality. Regarding the season effect, the best results were obtained in spring and summer, and the worst in autumn and winter. Chapter 4 dealt with the analysis of sources of variation of sensory traits and quality index of Trentingrana cheese. Fixed effects included in the model were dairy factory, and year and season of production All factors significantly affected the studied traits, with the exception of interactions between dairy and season of production for texture and external aspect, and between year and season of production for odor. Dairy was the most important source of variation for visually assessed traits and quality index, whereas the year of production was the most important for flavor attributes. The latter traits were always highly correlated among them and with the quality index. Correlations among visually assessed attributes, between them and flavor attributes, and between them and the quality index were more erratic.