Bioelectrical Impedance Analysis to Estimate Lipid Content in Atlantic Salmon Parr as Influenced by Temperature, PIT Tags, and Instrument Precision and Application in Field Studies

Abstract

The objective of this study was to investigate the effectiveness of bioelectrical impedance analysis (BIA) in field studies monitoring Atlantic Salmon Salmo salar parr, as related to temperature corrections, instrument precision, and passive integrated transponder (PIT) tags. Currently, BIA studies are restricted to laboratory settings where water temperature is controlled to decrease error in BIA predictions caused by fish body temperature. We compared models of predicted total and percent lipids with and without temperature corrections and found that temperature corrections reduced error caused by temperature. Without temperature corrections, an 8°C increase in temperature increased the predicted total lipids by 55%. After temperature corrections were added, the predicted total lipid only increased by 2.55%. Repeated measurements were collected on 40 salmon parr (56–115 mm FL) in four separate time trials (1 min, 1.5 h, 3 h, and 6 h), and we found that lipid content predictions between measurements were not significantly different; however, the variability within longer time trials was moderate (6.43% error). No significant differences were found in the predicted lipid value before or after PIT tags were removed from the body cavity, suggesting PIT tags do not affect BIA readings. On average, the difference between predicted total lipids after tag removal was 0.0023 and 0.02 g for 12.5-mm and 22-mm PIT tags, respectively. We also observed that increases in fish body temperature caused by handing resulted in increased variability in BIA estimates, indicating the need for temperature corrections.

Received March 10, 2014; accepted September 22, 2014

ACKNOWLEDGMENTS

We thank Danielle MacDonald, Michelle Charest, Stephanie Ratelle, and Paula Thoms for their help performing experiments. This project was funded by a Natural Sciences and Engineering Research Council of Canada HydroNet Strategic Network Grant to R.A.C. We would also like to thank our anonymous reviewer and Andrew Hafs for their insightful comments and suggestions during the initial iterations of this manuscript.