Meta‐analysis and critical review of trophic discrimination factors (Δ¹³C and Δ¹⁵N): Importance of tissue, trophic level and diet source

Abstract

Robustly quantifying dietary resource use and trophic position using stable isotopes requires accurate trophic discrimination factors (TDF; Δ13C and Δ15N for carbon and nitrogen, respectively), defined as the isotopic difference between consumer and diet. Early TDF studies converged on values of around 1.0‰ for Δ13C and 3.4‰ for Δ15N but more recent work indicates that TDF values may be more nuanced, depending on taxa, tissues, trophic level and diets. Yet, the relative importance of these factors remains unclear. Focusing on vertebrates (birds, fish, herptiles and mammals), we conducted a literature review of 279 studies that estimated TDF values and used a Bayesian framework to determine how tissue type, trophic level and diet source influence variation in Δ13C and Δ15N. Additionally, we reviewed 358 trophic ecology studies to determine if studies accounted for these factors during their TDF selection process. For Δ13C, vertebrates showed consistent patterns among tissue types (likely influenced by amino acid composition) and between trophic levels and diet sources (likely a result of dietary protein content and metabolic routing). Comparatively, for Δ15N, vertebrates showed considerable variation among tissue types and trophic levels, likely due to differences in tissue synthesis and physiological capabilities. Overall, Δ13C ranged from −5.1‰ to 9.1‰ and Δ15N from −3.3‰ to 9.7‰, underscoring that 1.0‰ for Δ13C and 3.4‰ for Δ15N are not universally appropriate. Moreover, both Δ13C and Δ15N varied by more than 9‰ within a single species and tissue type, demonstrating that using TDF values from the same, or similar, species may not be appropriate if diet and trophic level are not considered. Despite the importance of diet source on TDF values, most trophic ecology studies did not account for it. Further, most fish studies relied on literature review values that failed to account for tissue type, trophic level and diet source. To aid ecologists in diet and trophic assessments of vertebrates, we used our meta‐analysis to model taxon‐specific TDF estimates (mean ± SD) for each tissue type, trophic level and diet source combination. These more refined TDF values should improve ecological assessments that use stable isotopes. Read the free Plain Language Summary for this article on the Journal blog.

Body size predicts ontogenetic nitrogen stable-isotope (δ15N) variation, but has little relationship with trophic level in ectotherm vertebrate predators

“…Most recent studies that have evaluated the trophic level of organisms have acknowledged the sources of bias listed in the introduction (i.e. non-constant TDF 14 , 15 , 23 – 25 ), but have continued to use δ 15 N data to estimate trophic level, presumably because they assume that variation due to these factors in the field is trivial in relation to the effects of the differences between organisms and their foods, or because they believe that the biases will be subsumed in the variance components of Bayesian analyses. However, studies of intraspecific variation in crocodilians 16 , 36 and a large fish 37 indicate that physiological factors associated with size or growth rate, which are presumably related to metabolism, are more important than diet in determining the values of δ 15 N in body tissues.…”

“…There is a rich literature on laboratory-derived TDFs for species fed controlled diets, as well as meta-analyses of those studies (e.g. 15 ). Most laboratory studies, however, are conducted with short-lived, fast-growing organisms that are easily kept in captivity (e.g.…”

“…There has been debate about the reliability of using constant TDF to determine trophic position 19 – 22 . It is well recognized that many factors affect TDF, including food quality, metabolic paths, developmental stage, growth rate, body mass, temperature, and sex 14 , 15 , 23 – 25 . In controlled studies, TDF of captive Atlantic salmon have been related to growth rate 25 , which is ultimately dependent on body size in ectotherms 26 .…”

Patterns in tern trophic diversity in a region experiencing rapid climate change

“…The magnitude of these shifts was surprisingly large given that seabirds often show relatively consistent diet over short time scales (Woo et al 2008). One common tern chick, for example, showed a Δδ 15 N of 2.2‰ over a nine-day period, suggesting a change of nearly one trophic level in diet using the often-assumed trophic discrimination factor of 3.4‰ (but see Stephens et al 2023 for a discussion of variation in trophic discrimination factors).…”

Enameloid‐bound δ15N reveals large trophic separation among Late Cretaceous sharks in the northern Gulf of Mexico

“…In modern food webs, the N isotope ratio (δ 15 N) is widely used to infer trophic level, integrated diet, and spatiotemporal variability therein (e.g., Ferreira et al., 2017; Malpica‐Cruz et al., 2012). Modern shark tissues increase in δ 15 N by approximately 2.0‰–5.5‰ relative to their prey (Hussey et al., 2010; Kim et al., 2012; Zeichner et al., 2017), depending on taxon, tissue‐type sampled, food source, and physiological status (i.e., age, size, inter‐individual preferences; e.g., Gorokhova, 2018; Lübcker et al., 2020; Stephens et al., 2023). This increase in δ 15 N with each trophic level is termed the trophic discrimination factor (TDF) and is the result of excretion of 14 N‐rich (low‐δ 15 N) metabolic waste products (e.g., urea, trimethylamine‐N‐oxide, uric acid), leaving animals with 15 N‐rich (high‐δ 15 N) tissues (Minagawa & Wada, 1984).…”