Utilization of carbon isotope enrichments (δ13C) of alkanes as faecal markers to estimate diet composition of goats fed with heathland vegetation.

Abstract: This study aimed to evaluate the possible utilization of carbon isotope enrichments (δ13C) of n-alkanes as faecal markers for estimating diet composition of goats fed with diets composed of different proportions of browse (Erica umbellata, Erica cinerea, Calluna vulgaris, Erica arborea, and Ulex gallii) and herbaceous feeds (Lolium perenne and Trifolium repens). Diet composition was estimated from δ13C of n-alkanes (C27 and C29–C33) in plant species and faeces alone and/or their combination with alkanes, long-chain fatty acids and long-chain alcohols, using least square optimization procedures. Large variation between plant species in their δ13C profiles was observed with the level of enrichment varying from −31.20‰ to −38.40‰ and tending to be higher for the odd-chain δ13C compared to even-chain ones (−34.12±1.90‰ vs. −35.13±1.39‰, respectively). Results suggest that δ13C values of alkanes provided different discriminatory information to that given by other markers. No differences were found between the faecal recovery of 13C and 12C in individual alkanes, showing evidence that no differential degradation in the gastrointestinal tract occurred between isotopes for any of the alkanes. These results indicate that δ13C values of alkanes can be used without the need to apply any faecal correction factor as they are relative values. Accuracy of diet composition estimates based on δ13C data alone was lower (P<0.05) than when using alkanes, LCFA and LCOH. Moreover, data showed that the inclusion of δ13C in the input data used in diet composition calculations resulted in lower (P<0.05) accuracy of the estimates in several combinations. Results indicate limitations on the use of δ13C values of alkanes, possibly because differences between plant species in their profiles may have been not sufficiently wide to obtain accurate estimates of diet composition. [Copyright &y& Elsevier]