Are you really what you eat? Stomach content analysis and stable isotope ratios do not uniformly estimate dietary niche characteristics in three marine predators.

Calculation of dietary niche characteristics using stable isotopes has become a popular approach to understand the functional role of taxa across food webs. An underlying assumption of this approach is that stable isotopes accurately reflect the dietary breadth of a species over a temporal duration defined by tissue-specific isotopic turnover rates. In theory, dietary niche estimates derived from fast turnover rate tissues (e.g., blood plasma and liver) may augment stomach content-derived estimates more agreeably than slower turnover rate tissues (e.g., muscle or fin). We tested this hypothesis by comparing commonly used dietary niche estimates derived from stomach contents (niche: Levins', Shannon-Wiener's, and Smith's), with those estimated using stable isotopes [niche: standard ellipse area (SEA), convex hull total area (TA), theta (θ), and ellipse eccentricity (E)] of liver and muscle tissue. Model species were three large-bodied sharks: white (Carcharodon carcharias), dusky (Carcharhinus obscurus), and scalloped hammerhead (Sphyrna lewini). Within-technique comparisons for niche and niche metrics (i.e., SEA vs. TA) were often correlated; however, we did not observe any statistically significant correlations between niche and liver/muscle tissue niche (i.e., Levins' vs. SEA). We conclude that niche and niche do not provide comparable estimates of dietary niche, at least for the three predator species examined. This fundamental discrepancy highlights technique-specific limitations to estimating organismal dietary niche and identifies a need for the use of clearly defined niche metrics, i.e., the standardized use and reporting of the term isotopic niche as proposed by Newsome et al. (Front Ecol Environ 5:429-436, 2007). Finally, further investigation into the factors underpinning niche is required to better contextualize this popular ecological metric when compared to niche.