Interspecific and nutrient-dependent variations in stable isotope fractionation: experimental studies simulating pelagic multitrophic systems.

Stable isotope signatures of primary producers display high inter- and intraspecific variation. This is assigned to species-specific differences in isotope fractionation and variable abiotic conditions, e.g., temperature, and nutrient and light availability. As consumers reflect the isotopic signature of their food source, such variations have direct impacts on the ecological interpretation of stable isotope data. To elucidate the variability of isotope fractionation at the primary producer level and the transfer of the signal through food webs, we used a standardised marine tri-trophic system in which the primary producers were manipulated while the two consumer levels were kept constant. These manipulations were (1) different algal species grown under identical conditions to address interspecific variability and (2) a single algal species cultivated under different nutrient regimes to address nutrient-dependent variability. Our experiments resulted in strong interspecific variation between different algal species (Thalassiosira weissflogii, Dunaliella salina, and Rhodomonas salina) and nutrient-dependent shifts in stable isotope signatures in response to nutrient limitation of R. salina. The trophic enrichment in (15)N and (13)C of primary and secondary consumers (nauplii of Acartia tonsa and larval herring) showed strong deviations from the postulated degree of 1.0 per thousand enrichment in delta(13)C and 3.4 per thousand enrichment in delta(15)N. Surprisingly, nauplii of A. tonsa tended to keep "isotopic homeostasis" in terms of delta(15)N, a pattern not described in the literature so far. Our results suggest that the diets' nutritional composition and food quality as well as the stoichiometric needs of consumers significantly affect the degree of trophic enrichment and that these mechanisms must be considered in ecological studies, especially when lower trophic levels, where variability is highest, are concerned.