The role of species and geography in the elemental profiles of farm-raised shrimp from Indonesia.

Abstract

Elemental profiling is being explored as a traceability tool in many seafood products. However, the extent that elemental profiling can be used at finer geographical scales in cultured shrimp is unknown. Additionally, few studies have included multiple species in the same discriminant models, which would be useful in applications where one species is common, and the other is not. Here, elemental profiling was used to discern the provenance of black tiger shrimp Penaeus monodon and whiteleg shrimp Litopenaeus vannamei from the regions of North Kalimantan, Sulawesi Seletan, and Aceh in Indonesia. ICP-MS was used to determine elemental concentrations of 41 elements in shrimp muscle tissue and was the basis for multivariate and univariate statistical analyses. A MANOVA showed that multivariate differences exist in regions and between species of shrimp sampled. Univariate comparisons were utilized after the significance of the MANOVA and showed that 19 of the 24 elements above detection limits had significant differences. Classification via random forest was used to access the ability to discern, region, species, and region × species group combinations. The lowest model accuracy was the region × species combinations at 78.9%, while the highest accuracy was species irrespective of geographical origin at 93.59%. Elements that were routinely important in classification included As, B, Ba, Li, Na, Rb, Se, and Zn. Elemental profiles of white leg shrimp and black tiger shrimp are varied and potentially should not be used in the same classification models. Altogether, these results suggest that elemental profiling of farmed shrimp at finer geographic scales needs refinement as a traceability tool.