Assessing the potential of amino acid δ <sup>13</sup> C and δ <sup>15</sup> N analysis in terrestrial and freshwater ecosystems
Alexi C. Besser, Emma A. Elliott Smith, Seth D. Newsome
Abstract
Abstract Understanding the structure and dynamics of food webs requires accurate estimates of energy flow among organisms. Bulk tissue carbon (δ 13 C) and nitrogen (δ 15 N) isotope analysis is often used to this end; however, the limitations of this technique can outweigh the benefits. The isotope analysis of individual amino acids is being increasingly employed to trace energy flow and estimate consumer trophic level. Central to this compound‐specific approach are the concepts of essential amino acid (AA ESS ) δ 13 C fingerprinting and amino acid (AA) δ 15 N β ‐values, both of which have been understudied and are poorly constrained in terrestrial and freshwater producers. We present AA ESS δ 13 C data for 112 terrestrial and freshwater producers collected from two aridland habitats in the northern Chihuahuan Desert (New Mexico, USA) and AA δ 15 N data for a subset ( n = 28) of these samples. We characterized AA ESS δ 13 C fingerprints by performing linear discriminant analysis on the δ 13 C values of isoleucine, leucine, lysine, phenylalanine, threonine and valine for four producer groups—C 3 plants, C 4 plants, CAM plants and filamentous green algae. We explored potential biochemical mechanisms underlying these AA ESS δ 13 C fingerprints by calculating differences between the δ 13 C values of AA ESS products and their AA precursors. This allowed us to estimate and compare isotopic discrimination for specific AA ESS synthesis pathways across producer groups. We found a near‐perfect separation of AA ESS δ 13 C fingerprints among producer groups; all groups reclassified with >95% success within our multivariate framework. We also found varied isotopic discrimination for specific AA ESS synthesis pathways among producer groups. Contrary to previous studies, we found no differences in β ‐values between terrestrial C 3 and C 4 plants for any trophic‐source AA pairing. Furthermore, we found that Lys δ 15 N values were less variable and more closely related to bulk tissue δ 15 N values than Phe δ 15 N values in terrestrial and freshwater producers. Synthesis . We conclude that AA ESS δ 13 C fingerprints are a higher resolution tracer for freshwater food webs, where instream algae have overlapping bulk tissue δ 13 C values with terrestrial C 3 plants. Additionally, β Glx‐Lys and β Pro‐Lys are the best for AA δ 15 N‐based consumer trophic‐level estimates in freshwater food webs containing both terrestrial and aquatic resources.