Distance measures and optimization spaces in quantitative fatty acid signature analysis

Measuring repeatability of compositional diet estimates: An example using quantitative fatty acid signature analysis

By measuring the temporal consistency, or repeatability, in the diets of predators, we can gain a better understanding of the degree of individual specialization in resource utilization and implications for predator-prey interactions, population dynamics, and food web structure. To measure repeatability, we require repeated diet estimates of individuals over time, such as those derived from quantitative fatty acid signature analysis (QFASA), a popular diet estimation technique. However, diet estimates are often lengthy compositional vectors with many zeros, as some prey will not be consumed by all individuals, precluding the use of previously proposed measures of repeatability. In this paper, we propose a novel approach for inferring repeatability for multivariate data and, in particular, compositional diet estimates. We extend the commonly used measure of repeatability for univariate data to the multivariate compositional setting by utilizing the mean squares obtained from a nonparametric multivariate analysis of variance, and an appropriate choice of statistical distance. Our measure and its extension are compatible with both balanced and unbalanced data sets. Associated confidence intervals via nonparametric bootstrapping are also developed for the case of QFASA diet estimates that incorporate both sampling error and measurement error, where the latter error arises because the diets of predators are estimated. Simulation study results suggest that for practical levels of repeatability, our methods yield confidence intervals with the desired coverage probability even when the sample size relative to the dimension of the data (i.e., number of prey species eaten) is small. We tested our methods using QFASA diet estimates for free-ranging Northwest Atlantic grey seals. Given the importance of understanding how predator diets vary over time and space, our method may find broad application to other compositional diet estimates, including those derived from the stomach or fecal contents, and stable isotope analyses.

Summer/fall diet and macronutrient assimilation in an Arctic predator

Free-ranging predator diet estimation is commonly achieved by applying molecular-based tracers because direct observation is not logistically feasible or robust. However, tracers typically do not represent all dietary macronutrients, which likely obscures resource use as prey proximate composition varies and tissue consumption can be specific. For example, polar bears (Ursus maritimus) preferentially consume blubber, yet diets have been estimated using fatty acids based on prey blubber or stable isotopes of lipid-extracted prey muscle, neither of which represent both protein and lipid macronutrient contributions. Further, additional bias can be introduced because dietary fat is known to be flexibly routed beyond short-term energy production and storage. We address this problem by simultaneously accounting for protein and lipid assimilation using carbon and nitrogen isotope compositions of lipid-containing prey muscle and blubber to infer summer/fall diet composition and macronutrient proportions from Chukchi Sea polar bear guard hair (n = 229) sampled each spring between 2008 and 2017. Inclusion of blubber (85-95% lipid by dry mass) expanded the isotope mixing space and improved separation among prey species. Ice-associated seals, including nutritionally dependent pups, were the primary prey in summer/fall diets with lower contributions by Pacific walruses (Odobenus rosmarus) and whales. Percent blubber estimates confirmed preferential selection of this tissue and represented the highest documented lipid assimilation for any animal species. Our results offer an improved understanding of summer/fall prey macronutrient usage by Chukchi Sea polar bears which likely coincides with a nutritional bottleneck as the sea ice minimum is approached.

Fatty acid profiles of feeding and fasting bears: estimating calibration coefficients, the timeframe of diet estimates, and selective mobilization during hibernation

Accurate information on diet composition is central to understanding and conserving carnivore populations. Quantitative fatty acid signature analysis (QFASA) has emerged as a powerful tool for estimating the diets of predators, but ambiguities remain about the timeframe of QFASA estimates and the need to account for species-specific patterns of metabolism. We conducted a series of feeding experiments with four juvenile male brown bears (Ursus arctos) to (1) track the timing of changes in adipose tissue composition and QFASA diet estimates in response to a change in diet and (2) quantify the relationship between consumer and diet FA composition (i.e., determine "calibration coefficients"). Bears were fed three compositionally distinct diets for 90-120 days each. Two marine-based diets were intended to approximate the lipid content and composition of the wild diet of polar bears (U. maritimus). Bear adipose tissue composition changed quickly in the direction of the diet and showed evidence of stabilization after 60 days. During hibernation, FA profiles were initially stable but diet estimates after 10 weeks were sensitive to calibration coefficients. Calibration coefficients derived from the marine-based diets were broadly similar to each other and to published values from marine-fed mink (Mustela vison), which have been used as a model for free-ranging polar bears. For growing bears on a high-fat diet, the temporal window for QFASA estimates was 30-90 days. Although our results reinforce the importance of accurate calibration, the similarities across taxa and diets suggest it may be feasible to develop a generalized QFASA approach for mammalian carnivores.

Are tissue samples obtained via remote biopsy useful for fatty acid-based diet analyses in a free-ranging carnivore?

Fundamental knowledge on free-ranging animals has been obtained through capture-based studies; however, these may be logistically intensive, financially expensive, and potentially inconsistent with local cultural values. Genetic mark–recapture using remote tissue sampling has emerged as a less invasive alternative to capture-based population surveys but provides fewer opportunities to collect samples and measurements for broader ecological studies. We compared lipid content, fatty acid (FA) composition, and diet estimates from adipose tissue of polar bears (Ursus maritimus) obtained from two collection methods: remote biopsies (n = 138) sampled from helicopters and hunter-collected tissue (n = 499) from bears harvested in Davis Strait and Gulf of Boothia, Nunavut, 2010 – 2018. Lipid content of adipose tissue was lower in remote biopsies than harvest samples likely because remote biopsies removed only the outermost layer of subcutaneous tissue, rather than the more metabolically dynamic innermost tissue obtained from harvest samples. In contrast, FA composition was similar between the two collection methods with relatively small proportional differences in individual FAs. For diet estimates in Davis Strait, collection method was not a predictor of prey contribution to diet. In Gulf of Boothia, collection method was a predictor for some prey types, but the differences were relatively minor; the rank order of prey types was similar (e.g., ringed seal; Pusa hispida was consistently the primary prey in diets) and prey proportions differed by < 6% between the collection methods. Results from both methods showed that diets varied by geographic area, season, year, age class, and sex. Our study demonstrates that adipose tissue from remote biopsy provides reliable estimates of polar bear diet based on FA analysis and can be used to monitor underlying ecological changes in Arctic marine food webs.

Review of Estimating Trophic Relationships by Quantitative Fatty Acid Signature Analysis

The dynamic predator–prey relations in the food web are vital for understanding the function and structure of ecosystems. Dietary estimation is a research hotspot of quantitative ecology, providing key insights into predator–prey relationships. One of the most promising approaches is quantitative fatty acid signature analysis (QFASA), which is the first generation of statistical tools to estimate the quantitative trophic predator–prey relationships by comparing the fatty acid (FA) signatures among predators and their prey. QFASA has been continuously widely applied, refined and extended since its introduction. This article reviewed the research progress of QFASA from development and application. QFASA reflects the long-term diet of predator, and provides the quantitative dietary composition of predator, but it is sensitive to the metabolism of predator. The calibration coefficients (CCs) and the FA subset are two crucial parameters to explain the metabolism of predators, but the incorrect construction or improper use of CCs and the FA subset may cause bias in dietary estimation. Further study and refinement of the QFASA approach is needed to identify recommendations for which CCs and subsets of FA work best for different taxa and systems.

The critical importance of experimentation in biomarker-based trophic ecology

Fatty acids are commonly used as biomarkers for making inferences about trophic relationships in aquatic and soil food webs. However, researchers are often unaware of the physiological constraints within organisms on the trophic transfer and modification of dietary biomarkers in consumers. Fatty acids are bioactive molecules, which have diverse structures and functions that both complicate and enhance their value as trophic tracers. For instance, consumers may synthesize confounding non-dietary sourced markers from precursor molecules, and environmental conditions also affect fatty acid composition. There is a vital need for more research on the uptake and transfer of trophic biomarkers in individual organisms in order to advance the field and make meaningful use of these tools at the scale of populations or ecosystems. This special issue is focused on controlled feeding experiments on a diverse taxonomic breadth of model consumers from freshwater, marine and soil ecosystems with a goal of creating a more integrated understanding of the connection between consumer physiology and trophic ecology. This article is part of the theme issue 'The next horizons for lipids as 'trophic biomarkers': evidence and significance of consumer modification of dietary fatty acids'.

Unlocking the power of fatty acids as dietary tracers and metabolic signals in fishes and aquatic invertebrates

Determining the transfer and transformation of organic matter in food webs is a fundamental challenge that has implications for sustainable management of ecosystems. Fatty acids (FA) offer a potential approach for resolving complex diet mixtures of organisms because they provide a suite of molecular tracers. Yet, uncertainties in the degree of their biochemical modification by consumers, due to selective retention or metabolism, have limited their application. Here, we consolidated 316 controlled feeding studies of aquatic ectotherms (fishes and invertebrates) involving 1404 species-diet combinations to assess the degree of trophic modification of FA in muscle tissue. We found a high degree of variability within and among taxa in the %FA in consumer muscle tissue versus %FA in diet regression equations. Most saturated FA had weak relationships with the diet (r² < 0.30) and shallow slopes (m < 0.30), suggesting a lack of retention in muscle when fed in increasing amounts. Contrarily, several essential FA, including linoleic (18:2n-6) and α-linolenic acid (18:3n-3), exhibited significant relationships with the diet (m > 0.35, r² > 0.50), suggesting supply limitations and selective retention in muscle by consumers. For all FA, relationships strengthened with increasing taxonomic specificity. We also demonstrated the utility of new correction equations by calculating the potential contributions of approximately 20 prey items to the diet of selected species of generalist fishes using a FA mixing model. Our analyses further reveal how a broad range of fishes and invertebrates convert or store these compounds in muscle tissue to meet physiological needs and point to their power in resolving complex diets in aquatic food webs. This article is part of the theme issue 'The next horizons for lipids as 'trophic biomarkers': evidence and significance of consumer modification of dietary fatty acids'.

Comparison of Bayesian and numerical optimization-based diet estimation on herbivorous zooplankton

Consumer diet estimation with biotracer-based mixing models provides valuable information about trophic interactions and the dynamics of complex ecosystems. Here, we assessed the performance of four Bayesian and three numerical optimization-based diet estimation methods for estimating the diet composition of herbivorous zooplankton using consumer fatty acid (FA) profiles and resource library consisting of the results of homogeneous diet feeding experiments. The method performance was evaluated in terms of absolute errors, central probability interval checks, the success in identifying the primary resource in the diet, and the ability to detect the absence of resources in the diet. Despite occasional large inconsistencies, all the methods were able to identify the primary resource most of the time. The numerical optimization method QFASA using χ²(QFASA-CS) or Kullback--Leibler (QFASA-KL) distance measures had the smallest absolute errors, most frequently found the primary resource, and adequately detected the absence of resources. While the Bayesian methods usually performed well, some of the methods produced ambiguous results and some had much longer computing times than QFASA. Therefore, we recommend using QFASA-CS or QFASA-KL. Our systematic tests showed that FA models can be used to accurately estimate complex dietary mixtures in herbivorous zooplankton. This article is part of the theme issue 'The next horizons for lipids as 'trophic biomarkers': evidence and significance of consumer modification of dietary fatty acids'.

Diversity and specificity of lipid patterns in basal soil food web resources

Soil food webs are important drivers for key ecological functions in terrestrial systems such as carbon and nutrient cycling. However, soil food web models generally lack quantitative data, mainly due to the shortage in high-throughput methods to describe energy flows. In marine environments, multivariate optimization models (Quantitative Fatty Acid Signature Analysis) and Bayesian approaches (source-tracking algorithm) were established to predict the proportion of predator diets using lipids as tracers. A premise for the application of such models to soil systems is to acquire the fatty acid pattern of a broad range of resources and to reveal potential overlap in their signatures. We present a comprehensive comparison of lipid pattern across widespread taxa of plants (leaves and roots, n = 48), algae (n = 59), fungi (n = 60), and bacteria (n = 62) as basal food web resources. Lipid profiles from microorganisms and algae were assessed from laboratory cultures, whereas plant tissue was derived from an arable field. A lipid library was constructed and multivariate data analyses (hierarchical clustering, nMDS) was used to assess the extent of separation in lipid pattern by species or resource type. The performance of the lipid library was tested by leave-one-prey-out (LOPO) analysis, giving the distinctiveness of the resource (prey) groups. Fungi and plant leaves were correctly assigned based on their lipid pattern with more than 98%, while plant roots and bacteria achieved 88 and 85%, respectively. However, algae were only correctly classified by 60%, pointing to a bias in the herbivore food chain. Fatty acids most important for separation of algae and plant leaves were of the omega 3 type, i.e. 16:3ω3 and 18:3ω3. In plant roots 18:1ω9 was most important, whereas bacteria were distinguished predominantly by methyl-branched fatty acids. Overall, the lipid pattern of major soil food web resources are sufficiently differentiated to allow for qualitative (biomarker) analyses as well as quantitative modelling, yet with precaution in the case of algae.

High contributions of sea ice derived carbon in polar bear (Ursus maritimus) tissue

Polar bears (Ursus maritimus) rely upon Arctic sea ice as a physical habitat. Consequently, conservation assessments of polar bears identify the ongoing reduction in sea ice to represent a significant threat to their survival. However, the additional role of sea ice as a potential, indirect, source of energy to bears has been overlooked. Here we used the highly branched isoprenoid lipid biomarker-based index (H-Print) approach in combination with quantitative fatty acid signature analysis to show that sympagic (sea ice-associated), rather than pelagic, carbon contributions dominated the marine component of polar bear diet (72–100%; 99% CI, n = 55), irrespective of differences in diet composition. The lowest mean estimates of sympagic carbon were found in Baffin Bay bears, which were also exposed to the most rapidly increasing open water season. Therefore, our data illustrate that for future Arctic ecosystems that are likely to be characterised by reduced sea ice cover, polar bears will not only be impacted by a change in their physical habitat, but also potentially in the supply of energy to the ecosystems upon which they depend. This data represents the first quantifiable baseline that is critical for the assessment of likely ongoing changes in energy supply to Arctic predators as we move into an increasingly uncertain future for polar ecosystems.

Simultaneous estimation of diet composition and calibration coefficients with fatty acid signature data

Knowledge of animal diets provides essential insights into their life history and ecology, although diet estimation is challenging and remains an active area of research. Quantitative fatty acid signature analysis (QFASA) has become a popular method of estimating diet composition, especially for marine species. A primary assumption of QFASA is that constants called calibration coefficients, which account for the differential metabolism of individual fatty acids, are known. In practice, however, calibration coefficients are not known, but rather have been estimated in feeding trials with captive animals of a limited number of model species. The impossibility of verifying the accuracy of feeding trial derived calibration coefficients to estimate the diets of wild animals is a foundational problem with QFASA that has generated considerable criticism. We present a new model that allows simultaneous estimation of diet composition and calibration coefficients based only on fatty acid signature samples from wild predators and potential prey. Our model performed almost flawlessly in four tests with constructed examples, estimating both diet proportions and calibration coefficients with essentially no error. We also applied the model to data from Chukchi Sea polar bears, obtaining diet estimates that were more diverse than estimates conditioned on feeding trial calibration coefficients. Our model avoids bias in diet estimates caused by conditioning on inaccurate calibration coefficients, invalidates the primary criticism of QFASA, eliminates the need to conduct feeding trials solely for diet estimation, and consequently expands the utility of fatty acid data to investigate aspects of ecology linked to animal diets.