Application of stable isotope analysis to study temporal changes in foraging ecology in a highly endangered amphibian

Abstract

Background: Understanding dietary trends for endangered species may be essential to assessing the effects of ecological disturbances such as habitat modification, species introductions or global climate change. Documenting temporal variation in prey selection may also be crucial for understanding population dynamics. However, the rarity, secretive behaviours and obscure microhabitats of some endangered species can make direct foraging observations difficult or impossible. Furthermore, the lethality or invasiveness of some traditional methods of dietary analysis (e.g. gut contents analysis, gastric lavage) makes them inappropriate for such species. Stable isotope analysis facilitates non-lethal, indirect analysis of animal diet that has unrealized potential in the conservation of endangered organisms, particularly amphibians.

Methodology/findings: I determined proportional contributions of aquatic macroinvertebrate prey to the diet of an endangered aquatic salamander Eurycea sosorum over a two-year period using stable isotope analysis of (13/12)C and (15/14)N and the Bayesian stable isotope mixing model SIAR. I calculated Strauss' dietary electivity indices by comparing these proportions with changing relative abundance of potential prey species through time. Stable isotope analyses revealed that a previously unknown prey item (soft-bodied planarian flatworms in the genus Dugesia) made up the majority of E. sosorum diet. Results also demonstrate that E. sosorum is an opportunistic forager capable of diet switching to include a greater proportion of alternative prey when Dugesia populations decline. There is also evidence of intra-population dietary variation.

Conclusions/significance: Effective application of stable isotope analysis can help circumvent two key limitations commonly experienced by researchers of endangered species: the inability to directly observe these species in nature and the invasiveness or lethality of traditional methods of dietary analysis. This study illustrates the feasibility of stable isotope analysis in identifying preferred prey species that can be used to guide conservation management of both wild and captive food sources for endangered species.

Figure 1. Eliza Spring, Zilker Park, Austin (Travis County), Texas and Transect Lines for Invertebrate Sampling.

The lower dam of Barton Springs Pool, which hosts another population of Eurycea sosorum, is visible in the distance. Arrow marks pipe where water exits Eliza Spring. Dashed lines indicate transects along which macroinvertebrate samples were taken.

Figure 2. Mean ±1SE prey density and Eurycea sosorum density from Eliza Spring.

N = 9 invertebrate samples per sampling date. Higher SE reflects more patchy spatial distribution of prey between sampling points. Only results for the three invertebrate species that met the criteria outlined in Methods are shown.

Figure 3. Stable isotope biplots of individual Eurycea sosorum and mean prey items.

2.31‰ is subtracted from each salamander δ¹⁵N‰ value to reflect isotopic discrimination. Prey stable isotope values are shown as study-wide means ±1SD.

Figure 4. Mixing model estimated contributions of amphipods, planarians and chironomids to the diet of Eurycea sosorum.

Bayesian credible intervals show estimated contributions of each prey item to the diet of E. sosorum derived from the stable isotope mixing model SIAR. Relative abundance of prey uses same scale on Y-axis.

Figure 5. Strauss’ linear electivity index for amphipods, planarians and chironomids in the diet of adult Eurycea sosorum.

Values near +1 indicate high selection despite low availability in the environment (strong electivity), values near zero indicate foraging in proportion to availability in the environment (no electivity), and values near −1 indicate that prey is not selected despite high availability in the environment (avoidance).