Abstract
Existing data conflict as to the capability of stable isotope analysis (SIA) for distinguishing carbon pathways in stream ecosystems. I compiled 803 published measurements of δ13C for allochthonous litter and both lotic attached algae and consumers to search for general patterns that transcend individual studies. Half of the fishes and 30% of the invertebrates exhibited δ13C values outside the range that could be ascribed to allochthonous litter. The enormous variability in attached algal δ13C from −40 to −20‰ (mean ± SD = −29 ± 4) completely enveloped that observed for terrestrial leaf litter (mean = −28 ± 1‰). Therefore, for 50% of the fishes and 70% of the invertebrates it was impossible to discriminate between allochthonous and autochthonous carbon dependency. Faunal δ13C did not change in any quantitatively predictable manner with removal of riparian trees. Three conclusions are possible from this secondary analysis: (1) autotrophic pathways within forested headwaters are much more important to lotic foodwebs than would be suggested by their particulate inputs alone; (2) the great variability in attached algal δ13C may often preclude use of SIA for identifying carbon pathways in stream ecosystems; thus (3) the utility of carbon SIA in understanding anthropogenic alterations to the carbon budget of streams is presently minimal.