The nested structure of a scavenger community

Abstract

Scavenging is a widespread phenomenon in vertebrate communities which has rarely been accounted for, in spite of playing an essential role in food webs by enhancing nutrient recycling and community stability. Most studies on scavenger assemblages have often presented an oversimplified view of carrion foraging. Here, we applied for the first time the concept of nestedness to the study of a species-rich scavenger community in a forest ecosystem (Białowieza Primeval Forest, Poland) following a network approach. By analysing one of the most complete datasets existing up to now in a pristine environment, we have shown that the community of facultative scavengers is not randomly assembled but highly nested. A nested pattern means that species-poor carcasses support a subset of the scavenger assemblage occurring at progressively species-rich carcasses. This result contradicts the conventional view of facultative scavenging as random and opportunistic and supports recent findings in scavenging ecology. It also suggests that factors other than competition play a major role in determining community structure. Nested patterns in scavenger communities appear to be promoted by the high diversity in carrion resources and consumers, the differential predictability of the ungulate carcass types and stressful environmental conditions.

Figure 1

Bipartite graphs depicting scavenging relations (interactions) between the species of facultative scavengers (open circles) and ungulate carcasses (filled circles) of different origin: (a, b) predation, (c) natural causes other than predation and (d) human harvesting. Each line linking a scavenger species and a carcass indicates that that particular scavenger fed on that particular carcass. Carcasses are arranged from left to right in the order of decreasing number of scavenger species visiting them. All scavenger species are represented and also arranged from left to right in the order of decreasing number of occurrences on all carcasses.

Figure 2

Binary matrices representing (a) the real scavenger community and (b, c) two randomized communities (one replicate for each null model). A filled square indicates an observed presence of species j on carcass i. Numbers label carcasses (rows) and scavenger species (columns). Carcasses are arranged in the order of decreasing number of scavenger species visiting them, and scavenger species are ranked in the order of decreasing number of occurrences on carcasses, in a way that both minimizes unexpectedness. The line represents the isocline of perfect nestedness. The scavenger species and their corresponding rank number in the nested matrix (real data) are listed below.

Figure 3

Contribution to the nestedness of the scavenger community (mean±s.e., s.d.) of (a) different carcass types; (b) carcasses available in the cold versus warm season and (c) avian and mammalian scavengers.

Figure 4

Relationship between the contribution to community nestedness of each scavenger species and its degree, shown as percentage of carcasses scavenged upon. The continuous line represents the correlation for mammals (filled circles) and the broken line for birds (open circles).