Statistical analyses
We size-corrected TDFs of muscle and liver tissue with the attempt to adjust for the variation explained solely by growth (Fry and Arnold 1982). Thus, for all analyses, we used the residuals of the significant regression of TDFs and fish length. To analyze the variation in TDFs between weight classes, we conducted four different ANOVAs with subsequent Tukey´s posthoc comparisons using size-corrected Δ13C and size-corrected Δ15N, for the two tissue types muscle and liver. Additionally, we conducted four linear regressions with the size-corrected Δ13C and size-corrected Δ15N, for muscle and liver tissue respectively as dependent variables and SMR as the predictor in order to analyze the relationship between TDFs and SMR. We compared SMR between the different weight classes using an ANOVA with subsequent Tukey´s posthoc comparisons. The assumptions of normal distribution and homogeneities of variances were met for all parametric analyses. We used IBM SPSS (version 25) for statistical analyses.