Was that my meal? Uncertainty from source sampling period in diet reconstruction based on stable isotopes in a syngnathid fish

Stable isotopes can be used to reconstruct diet in fishes, an approach that is particularly useful in threatened or endangered species where lethal sampling needs to be avoided To apply this method stable isotopes need to be measured both in the consumers and dietary sources As turnover rates may largely differ across tissues, the reconstruction should consider the uncertainty associated with sampling timing for both consumers and dietary sources. The present study investigated for the first time the application of a time lag for the estimation of the trophic position and the dietary regime in the greater pipefish Syngnathus acus (Family Syngnathidae). For that, two isotopic (delta C-13 and delta N-15) data sets from dorsal fin-clipped tissues were used, including values in potential sources sampled (a) concomitantly with consumers (Unshifted approach) or (b) in the season sampled preceding consumer sampling (Shifted approach). The results attained indicate that the trophic position of the species was not affected (< 0.6% change) by shifting when analyzed for the annual dataset (global assessment). However, the trophic level estimated in the shifted treatment was ca. 6% higher than in the unshifted approach when the analysis was performed using a seasonal dataset (seasonal assessment including exclusively specimens collected in summer). Regarding the reconstruction of the dietary regime and in comparison to the unshifted approach, the results attained with shifted sampling revealed a higher preference for amphipods (43.7%), isopods (19.5%) and mysidaceans (12.1%), and a reduction in the reconstructed abundance for harpacticoid copepods (5.1%). These findings were similar to those attained for specimens and food web components collected in the same sampling period (seasonal dataset), both in females and males. This study demonstrates (1) the variability of estimates in the reconstruction of diet regimes depending on the data time shifting applied, and (2) the importance of considering adequate time lags in the assessment for syngnathid fishes.