Trophic analysis of tetrapod and arthropod species of the world

Trophic structures of birds and mammals:We studied the trophic composition of bird and mammal communities around the world, and their classification into six distinct guild configurations. Communities include 15265 species of terrestrial mammals (5272 species) and non-marine birds (9993 species) and were derived from distribution data from the IUCN Global Assessment distributional data for native ranges across a global grid of 1-degree pixels. Each species was classified into distinct guilds (plantivores, granivores, frugivores, nectivores, plant-invertivores, frug-invertivores, carnivores, specialist invertivores, and generalist invertivores) according to published diet data. For further details on methods implementation see. For our analysis, we aimed to characterize the trophic structure in the terrestrial realm, and thus excluded species that forage primarily in aquatic environments (i.e. fish diet higher than 30%). We quantified each species diet based on its percentage of primary consumption (i.e. plant sourced: plants, seeds, fruit, nectar, and pollen) and secondary consumption (i.e. animal sourced: invertebrates, vertebrate endotherms, vertebrate ectotherms, fish, carrion, and unknown vertebrates). We used this information to classify each species according to their trophic level either as primary consumers (i.e. more than 2/3 of diet consists of primary consumption), secondary consumers (i.e. more than 2/3 of diet consists of secondary consumption), and mixed consumers (i.e. between 1/3 and 2/3 of diet consists of primary or secondary consumption). Then, we calculated the proportion of species of each trophic level present in communities at each pixel. We used a Discriminant Analysis to assess the degree of differentiation in the trophic composition between guild configuration patterns (i.e. depauperate, boreal, temperate, semi-arid, seasonal tropical, and humid tropical), where the proportion (arcsine(√x) transformed) of primary, mixed, and secondary consumer species in each pixel (n=14563) served as response variables.Global trophic structure of tetrapods and arthropods:We studied the trophic composition of all terrestrial tetrapods and arthropods on the globe by extending our analysis to also include reptile, amphibian, and arthropod species. After excluding taxa feeding primarily from aquatic sources, our analysis encompassed 5237 species of mammals, 9271 birds, 8767 reptiles, 2477 amphibians, 1075219 arthropods. We classified species as primary, secondary, or mixed-consumers by associating numbers of species across taxa to the diets described in their natural history at the level of species for mammals and birds, family or lower for insects, order or lower for reptiles and amphibians, and order for other arthropods. If more than one diet was reported for a taxon, we aimed to find the sub-taxa and number of species exhibiting each diet. Otherwise, we assumed an even distribution of each diet across species in the taxon. When predominance of species with a diet was specified (e.g. descriptions like “most species are”), we assumed 65% to 95% of species—randomly assigned (see last paragraph for details on this and other sources of uncertainty)—exhibited such diet, while we assumed an even distribution of the remaining species across alternative diets unless otherwise specified. If more than one diet was reported within species—rather than across species—in a taxon, we specified the percentage of the total diet comprised by each diet type following the same rules described above. Reptile and amphibian diets were classified as herbivory, carnivory, and detritivory. Arthropod diets were classified as fungivory (i.e. feeding on hyphae or spores), phytophagy (i.e. feeding on any living tissue of plants including fruits), nectivory (i.e. feeding on sap, flower nectar, and pollen), predation, parasitism, ectoparasitism (i.e. feeding on blood or haemolymph), scavenging (feeding on animal carcasses), plant detritivory (feeding on death wood or leaves), and generalist saprophagy (feeding on unspecified rotting matter). We used diet information to calculate the percentage of primary and secondary-level consumption of species across taxa using the same thresholds as for birds and mammals. Phytophagy and nectivory were classified as primary-level consumption and predation, parasitism, and ectoparasitism as secondary-level consumption. We classified fungivory as secondary-level consumption because fungi are themselves primary consumers. Plant detritivory was classified as 100% to 50% primary-level consumption as it is unknown how much of this diet derives energy from the plant biomass or from fungi consuming the plant biomass. General detritivory and saprophagy was classified as 100% to 0% primary-level consumption as it is unknown how much if this diet derives energy from algal, bacterial, plant, fungal, or animal matter. We classified scavenging as secondary-level consumption as whether an animal prey is death or alive bears no importance to its trophic status as a consumer. We considered kleptoparasiting, dung eating, and honeydew consumption as secondary-level consumption (i.e. within scavenging) as the they are themselves products of animal consumers. For taxa with metamorphosis (e.g. anurans and holometabolous insects), we conducted a differential analysis of diet composition between developmental stages and average them. In many insect taxa the bulk of energetic consumption and biomass accumulation occurs in the larval stage, while adults exhibit an exclusive nectivorous diet providing the minimum necessary nourishment. Thus, for insect taxa with exclusively nectivorous adults, we limited adult diet weight from 10% to 50% when averaging with the larva diet. We classified species according to their trophic level either as primary consumers (i.e. more than 2/3 of diet consists of primary consumption), secondary consumers (i.e. more than 2/3 of diet consists of secondary consumption), and mixed consumers (i.e. between 1/3 and 2/3 of diet consists of primary or secondary consumption). We addressed uncertainty regarding the proportion of predominant diets across and within taxa, the proportion of primary consumption involved in general saprophagy and detritivory, and the energetic weight that adults represent in insect taxa with metamorphosis by randomizing these values in 1,000 trophic pyramids.