Lipid corrections in carbon and nitrogen stable isotope analyses in freshwater fishes of the Paraná river, Argentina.

LLAMAZARES VEGH, SABINA; ADOLFO GIL; ALEJANDRA V. VOLPEDO

Viña del MAr

Conferencia; 11th International Conference on the Applications of Stable Isotope Techniques to Ecological Studies (IsoEcol 2018); 2018

Universidad Andrés Bello,

Stable isotope analysis (SIA) coupled with mixingmodels are powerful tool for examining diet and food-web dynamics in aquaticecosystems. Nitrogen isotopes can be used to determine trophic position,increasing about 3·4? at each trophic level. Carbon stable isotopes can be usedto trace energy sources in food webs, because isotope values increase onlyslightly between diet and consumer (0?1?) during trophic transfers. Within anorganism, lipids are depleted in 13C relative to proteins andcarbohydrates, and variation in lipid content between species has the potentialto increase variability in δ13C values. Despite the potential forintroduced error, there is no consensus on the need to account for lipids in SIA.Several studies recommend accounting for lipids through chemical extractionprior to SIA or mathematical normalization when the lipid content in sample tissueis high (i.e., C:N >3.5, lipid content>5%) in fish muscle. Whilenormalization models have been tested for some aquatic organisms, there has beenno investigation of freshwater fauna of the Paraná river, the second largestriver in South America. In this study, effects of lipid extraction andmathematical correction methods for eight freshwater fishes were compared tobetter understand impacts of normalization approaches on carbon (δ13C)and nitrogen (δ15N) stable isotope data. Following Sweeting lipidextraction, δ13C values were higher for all species with a meanincrease of 0.36 ±0.23?. In contrast, there was a mean decrease in δ15N of 0.42 ±0.37. The lipid content rangedfrom 3.1% to 5.5% for Leporinus obtusidensand Cyphocharax voga,respectively. Only for C. voga theC:N >3.5 before treatment, extraction resulted in a decrease in C:N ratioand changes up to 1? in δ13C. For this species similar values of δ13Clipid-free were obtain using previous normalization models. The best fitrelationship for δ13C was δ13C lipid-free= 0.88×δ13Cbulk?3.00(R2=0.97, p<0.001).