Trace element composition of smolt scales from Atlantic salmon (Salmo salar L.), geographic variation between hatcheries

• Published in: Fisheries research Vol. 190; pp. 183 - 196
• Main Authors: Flem, Belinda, Moen, Vidar, Finne, Tor Erik, Viljugrein, Hildegunn, Kristoffersen, Anja Bråthen
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01-06-2017
• Subjects: Atlantic salmon; Geochemistry; Hydroxyapatite; Salmon management; Scale chemistry; Geochemistry; Scale chemistry; Atlantic salmon; Hydroxyapatite; Salmon management
• ISSN: 0165-7836; 1872-6763
• DOI: 10.1016/j.fishres.2017.02.010

•Sclerites of 1347 Atlantic salmon smolt scales from 18 hatcheries along Norway’s coastline are analysed by LA-ICP-MS.•Identification of hatchery of origin by 12 chemical elements was successful to 95.8%.•With the given natural conditions, Sr and Ba were the most important single elements for correct classification.•Comparing scale and floodplain sediment chemistry, 4 of 7 hatcheries are within a distinct area with high floodplain lead concentrations.
The escape of farmed Atlantic salmon is a problem for the Norwegian aquaculture industry. Escapes in all phases of the life cycle of the farmed fish create situations unwanted by authorities and industry, as well as with those depending on the wild Atlantic salmon for commercial or recreational fishing. As a contribution to develop a dependable and feasible method of linking escaped salmon to their hatchery, the chemical variation of 12 elements in sclerites of scale from smolt from hatcheries located all along the coast have been analysed by laser ablation inductively coupled plasma mass spectroscopy (LA-ICP-MS). The survey comprised data on B, Ba, Fe, K, Mg, Mn, Na, Pb, S, Sr, U and Zn in 1347 scales from 24 groups of smolt collected from 18 hatcheries, the largest survey with regard to statistical and geographical variation conducted to date. A majority vote between the result of four different classification methods (linear discriminant analysis, Naïve Bayes, k-nearest neighbour and random forest) misplaced only 55 scales (4%) to group of origin. Sr and Ba are the two most important elements for group separation. Without Sr, the success of correct hatchery classification is reduced from 95.8% to 92.2%, and to 93.4% without Ba.
Chemical composition (same elements as in sclerites, except for S and U) of local floodplains’ sediments revealed elevated levels of Pb coinciding with most of the hatcheries where Pb in sclerites was above detection limit.
The other elements incorporated in the hydroxyapatite layer (HAP-layer) analysed for this study seems shifted by water treatment done in the hatcheries. Scale chemistry may provide a powerful tool to determine the origin of escaped farmed Atlantic salmon. However, further investigations should be performed before this method can be implemented in the Norwegian salmon management, particularly on two topics. These are the stability of trace elements incorporated in the HAP-layer after change of water environment, and whether the elemental profile from the freshwater hatchery is still recognizable after the salmon has spent more than 2 years in sea.