Abstract
This thesis aims to develop a method to investigate the ability of bulk stable isotope and compound specific stable isotope tools to verify whether dietary feed fatty acids, bovine milk water and major bovine milk fatty acids are conveying biogeochemical attribution of their production region. This technique opens new insights into milk regional authenticity identification.
A multiple linear regression (MLR) model based on New Zealand climatic parameters was employed to verify its ability to predict δ2H values of rainfall from milk production regions. The MLR model showed promise to predict δ2H values of regional rain, yielding a high correlation to actual rain δ2H values (R2 = 0.73, P<0.05). Subsequently the model’s generated estimates of δ2H values for precipitation of the milk production regions, revealed strong geographical correlation between bulk δ2H values and δ2H values of some bovine milk short chain fatty acids (SCFA) to long chain fatty acid (LCFA).
In order to understand which of the major fatty acids in milk powder have the potential discrimination capability to allow the provenancing of milk powder, the δ2H value of C4:0 (butyric), C14:0 (myristic), C16:0 (palmitic) and the δ2H value of bulk milk which showed the highest correlation to regional rainfall was selected and analysed by employing multivariate statistics. The δ2H values of these compounds were found to be capable of explaining 91% of the isotopic variation. The hydrogen isotopic compositions of these milk compounds were able to separate milk production regions across New Zealand.
Subsequent exploration into finding the isotopic link between farm drinking water, grass/feed and milk, revealed that bovine milk bulk and fatty acid hydrogen isotope composition carries isotopic attributions both from feed and local water. However the influence of regional water on the 2H composition of the milk seems to be more pronounced.
The influence of seasonal variability on milk was examined on milk powder samples sourced from Norway. Results indicate that the milk powder fatty acid δ2H values and fatty acid concentrations were influenced by the time of sampling throughout the year, while the bulk milk δ2H and δ13C values remained relatively consistent across the sampling period. This may suggest a presence of a region specific isotope variability that may further be explored by comparing it with variability patterns of other regions.
A preliminary assessment on the potential of δ2H from milk fatty acids and bulk milk for determination of an adulterant (an unknown milk powder) in milk powder was investigated. Multivariate statistics were used to quantify the level of adulteration, which was able to resolve differences at a 5% level of adulterant in an authentic sample.
This thesis explores the potential application of stable isotope analysis to the authentication of the provenance of New Zealand milk. The approach utilized in this study could be adopted in other milk producing regions, or applied to other milk products such as infant formula for authentication purposes.