Opportunistic top predators partition food resources in a tropical freshwater ecosystem

Abstract

The structure of food webs may be strongly influenced by the distribution of top predators in space and time. The Amazon biome is the only region in the world where four alligatorid species are known to occur in sympatry, and they attain high densities in some regions. As top predators with a diverse range of prey species occupying different trophic levels, their impact upon food webs should be substantial, but the degree to which crocodilians differ in their food sources, and potentially avoid competitive exclusion where they occur syntopically is not well understood. Although most crocodilians are considered generalist opportunistic predators that feed on any source of protein available in the environment, Amazonian crocodilians show broad differences in the proportions of prey items they consume. It is believed that these differences may in part reflect habitat use, but it is unknown to what extent they represent interspecific differences in prey preferences or are a direct function of habitat selection. Stable carbon isotope data (δ13C) of crocodilians and their potential prey were used to assess differences in reliance on terrestrial versus aquatic resources. These data were then placed in a spatial context using classified maps that reflect habitat types (headwater streams, mid‐order flooded‐forest streams and várzea floodplains) to elucidate whether dietary differences are explained by habitat selection or are more likely a reflection of prey preferences. We found evidence for differences in types of basal resources supporting these crocodilians. Mean δ13C values were highest in Paleosuchus trigonatus (Schneider's dwarf caiman, −25.7 ± 1.2‰), intermediate in Caiman crocodilus (spectacled caiman, −27.4 ± 1.2‰) and Paleosuchus palpebrosus (Cuvier's dwarf caiman, −27.7 ± 1.1‰) and lowest in Melanosuchus niger (black caiman, −29.9 ± 1.3‰). A progressive decrease in δ13C values of crocodilian tissues occurred from headwaters to floodplains, which most likely reflects a progressive increase in autochthonous over allochthonous inputs in lower reaches of streams. The shift from terrestrial to aquatic resources sustaining these sympatric predators mirrors their spatial distribution along this ecotone. However, after taking into account the habitat in which pairs of syntopic individuals of distinct species occurred, significant differences in δ13C values suggest that P. trigonatus and P. palpebrosus have different prey bases. Thus, despite being opportunistic predators, our results show that differences in crocodilian diets likely result from prey preferences and not only from habitat selection. These findings suggest that some species of crocodilians may be less generalist than traditionally thought and their influence on terrestrial or aquatic food webs might be species specific.

Body size is more important than diet in determining stable-isotope estimates of trophic position in crocodilians

“…This approach typically yields results in which baseline variation is not a major source of error in trophic position estimates 63 . Furthermore, if consumers ingest food from a number of distinct food chains, such as species that use aquatic and terrestrial food resources, supplementary information on the proportional contributions of those sources to the diet, such as that provided by δ 13 C data, is necessary to estimate trophic position based on δ 15 N. After correcting for this source of uncertainty by applying the proportional contributions of aquatic and terrestrial carbon resources to the crocodilian tissues 19 to calculate trophic position, we were confident of accounting for habitat variability in δ 15 N sources.…”

“…base are nitrogen stable isotope values of terrestrial and aquatic baselines (4.6‰ and 3.8‰, respectively); α is the proportional contribution of carbon from terrestrial origin (obtained from ref. 19 ); λ is the trophic level of the organisms used to estimate δ 15 N . terr .…”

“…Based on the mass of a prey item that a crocodilian of 100 cm SVL would ingest (which would be the largest possible size of the prey), we calculated the expected mass ( Prey.mass.exp ) of a prey item that a crocodilian of a given volume (SVL 3 ) would ingest. We estimated prey mass per crocodilian: To calculate Pj, we estimated the proportional contribution of each prey category within the diet of that crocodilian species based on Prey.mass.croc and aquatic (Aq.prop) vs. terrestrial (Terr.prop) contributions derived from δ 13 C 19 : …”

“…To calculate Pj, we estimated the proportional contribution of each prey category within the diet of that crocodilian species based on Prey.mass.croc and aquatic (Aq.prop) vs. terrestrial (Terr.prop) contributions derived from δ 13 C 19 :…”

“…As opportunistic predators 17 , 18 , crocodilians may strongly influence the structure of food webs in distinct ways by preying upon organisms occupying different trophic levels across aquatic and terrestrial systems 14 , 16 , 19 – 22 . In all species, crocodilians grow several orders of magnitude both in length and mass during their lifespan.…”