Study of the Interface between Wild Bird Populations and Poultry and Their Potential Role in the Spread of Avian Influenza

Abstract

Water birds play a crucial role in disseminating and amplifying avian influenza viruses (AIVs) in the environment. However, they may have limited interactions with domestic facilities, raising the hypothesis that other wild birds may play the bridging role in introducing AIVs into poultry. An ornithocoenosis study, based on census-transect and camera-trapping methods, was conducted in 2019 in ten poultry premises in northeast Italy to characterize the bird communities and envisage the species that might act as bridge hosts for AIVs. The data collected were explored through a series of multivariate analyses (correspondence analysis and non-metric multidimensional scaling), and biodiversity indices (observed and estimated richness, Shannon entropy and Pielou's evenness). The analyses revealed a high level of complexity in the ornithic population, with 147 censused species, and significant qualitative and quantitative differences in wild bird species composition, both in space and in time. Among these, only a few were observed in close proximity to the farm premises (i.e., Magpies, Blackbirds, Cattle Egrets, Pheasants, Eurasian Collared Doves, and Wood Pigeons), thus suggesting their potential role in spilling over AIVs to poultry; contrarily, waterfowls appeared to be scarcely inclined to close visits, especially during autumn and winter seasons. These findings stress the importance of ongoing research on the wild-domestic bird interface, advocating for a wider range of species to be considered in AIVs surveillance and prevention programs.

Keywords: HPAI; bridge hosts; camera trap; ornithocenosis; ornithological transects; spillover; wild birds; wild–domestic interface.

Figure 1

Farms’ locations included in the study. In grey: domestic and wild outbreaks reported in the study area during the HPAI H5N8 epidemic of 2017–2018.

Figure 2

Overall abundances of the observed wild birds are presented. (Left) points to abundance distributions per farming site; (right) points to abundance distributions per month. The middle, lower and upper hinges of the box plots represent the 50%, 25% and 75% quantiles, respectively; the black dots represent the outliers.

Figure 3

Correspondence analysis plots are presented: (a) association between sites and common species; (b) association between months and common species. The species are colored according to the Order to which they belong, and the size represents their absolute abundance. CA plot points’ positions reveal associations based on their angle and distance from the axes’ origin: same direction for positive, opposite for negative associations; shorter distances for even species distribution, longer for unequal abundances.

Figure 4

(a) Dendrogram and (b) nMDS according to the Bray–Curtis distance (stress: 0.09) are presented. Grey represents Cluster 1, yellow, Cluster 2, and blue, Cluster 3.

Figure 5

Diversity indices (Chao1, Shannon entropy and Pielou’s eveness) distributions for the three identified clusters are presented. The middle, lower and upper hinges of the box plots represent the 50%, 25% and 75% quantiles, respectively; the black dots represent the outliers.

Figure 6

Number of days in which the target species for passive surveillance of avian influenza (EFSA) were observed in more than one farm during the study period in each farm. The Great White Egret and the Peregrine Falcon were excluded from the figure as they were poorly represented. Data were aggregated according to the clusters identified by the nMDS method applied to the transects’ data.

Figure 7

Number of days in which the seven camera-trapped species (including Magpie among the target species) observed in at least five farms were camera-trapped during the study period. Data were aggregated according to the clusters identified by the nMDS method applied to the transects’ data.