Comparison of morphological and molecular methods to identify the diet of a generalist omnivore

Abstract

Abstract ContextEcologists need robust and effective methods to quantify the diet of animals. However, assessing dietary composition can be challenging because most animals are seldom observed eating, especially when studying rare or cryptic species. AimsMorphological analysis of scats has been extensively used previously, and recent advances in the accessibility of DNA barcoding techniques have also made molecular approaches a viable alternative for diet analysis from scats. We compared the results from two methods of scat analysis, to trial the use of contemporary approaches in scat analysis. MethodsIn the present study, we used morphological analysis and DNA barcoding of matter in scats to catalogue the diet of a generalist omnivore, the greater bilby (Macrotis lagotis Thylacomyidae), in the West Kimberley. The composition and diversity of diet items, as well as the taxonomic identification level, were compared between methods. Key resultsEach method provided complimentary results; morphological analysis uncovered the type of matter consumed (e.g. root, seed) and relative proportion of the total undigested content, whereas DNA barcoding could assign such matter to a taxon. Even though dietary DNA could be extracted from only 38% of scats, DNA barcoding identified a greater diversity of taxa in scats than did morphological analyses. Barcoding could detect the presence of highly-digestible items such as cossid moths (Cossidae) and spiders (Araneae). ConclusionsMorphological analysis was useful for quantifying relative abundance of diet categories; however, DNA barcoding detected a greater diversity of dietary items within scats. Despite the expense of DNA barcoding, the method can more accurately identify the taxa consumed, whereas morphology can greatly underestimate dietary species diversity. However, the technical requirements for performing DNA analysis make it expensive, while resource-limited field ecologists can generally perform morphological analysis with appropriate training. ImplicationsResearchers and land managers will benefit from using both approaches in concert to gain a robust understanding of the local bilby diet. However, the cost and limitations of DNA barcoding (particularly when dealing with degraded DNA) mean that this approach should only be employed when the quality of the genetic material within samples is suitable. We recommend conducting exploratory analysis using morphological analysis (potentially in the field), with follow-up DNA barcoding to detect highly digestible items in fresh scats.