Compound‐specific isotope analysis of amino acids as a new tool to uncover trophic chains in soil food webs

Food webs in soil differ fundamentally from those aboveground; they are based on inputs from both living plants via root exudates, and from detritus, which is a complex mixture of fungi, bacteria, and dead plant remains. Trophic relationships are difficult to disentangle due to the cryptic lifestyle of soil animals and inevitable microbial contributions to their diet. Compound‐specific isotope analysis of amino acids (AAs) is increasingly used to explore complex food webs. The combined use of AA δ¹³C and δ¹⁵N values is a promising new approach to disentangle trophic relationships since it provides independent but complementary information on basal resources, as well as the trophic position of consumers. We conducted a controlled feeding study in which we reconstructed trophic chains from main basal resources (bacteria, fungi, plants) to primary consumers (springtails, oribatid mites) and predators (gamasid mites, spiders). We analyzed dual compound‐specific isotope AA values of both resources and consumers. By applying an approach termed “stable isotope (¹³C) fingerprinting” we identified basal resources, and concomitantly calculated trophic positions using ¹⁵N values of trophic and source AAs in consumers. In the ¹³C fingerprinting analysis, consumers in general grouped close to their basal resources. However, higher than usual offsets in AA δ¹³C between diet and consumers suggest either gut microbial supplementation or the utilization of specific resource fractions. Identification of trophic position crucially depends on correct estimates of the trophic discrimination factor (TDFGₗᵤ‐Pₕₑ), which was close to the commonly applied value of 7.6‰ in primary consumers feeding on microbial resources, but considerably lower in arachnid predators (~2.4‰), presumably due to higher diet quality, excretion of guanine, and fluid feeding. While our feeding study demonstrates that dual compound‐specific AA analyses hold great promise in delineating trophic linkages among soil‐dwelling consumers and their resources, it also highlights that a “one‐size‐fits‐all” approach to TDFGₗᵤ‐Pₕₑ does not apply to soil food webs.