In recent years, in large parts of the Northern hemisphere there has been an increasing focus on enhancing animal welfare within the agricultural sector, encompassing poultry production (1, 2). To address these concerns, new production systems have been designed and implemented, providing animals with the opportunity to reside in environments that are more like their natural habitats and less restrictive. However, these innovative production systems, especially those employed in the poultry industry, can result in greater interaction between domestic and wild birds, including their feces (3–5).

Various species of wild birds, known as synanthropic birds, belonging to the families Columbidae, Corvidae, and Passeridae, have demonstrated a remarkable adaptation to exploit resources generated by human activities, such as food, water and shelter (6). Examples from these families that include the house sparrow (Passer domesticus), the tree sparrow (Passer montanus), European starlings (Sturnus vulgaris), and feral pigeons (Columba livia), can inhabit diverse environments created by humans, from urban areas to isolated farms. Some of these birds, especially the house sparrow easily enter production facilities, through small gaps in exterior walls often even when protected by bird nets (6). Thus, if individuals of such species are in close contact with poultry on one hand and wild bird species such as waterfowl that not usually enter enclosures or barns on the other hand, they could act as so-called bridge hosts in the transmission of pathogens.

In terms of risks in addition to abundance of farm birds, composition of the farm bird community could be important (7). The dilution effect hypothesis postulates that a higher biodiversity is linked to a lower prevalence of pathogens, as species-rich communities harbor individuals in which a specific pathogen cannot multiply to sufficient levels to transmit infection to new susceptible individuals. This reduces the overall success of pathogen transmission and, consequently, the prevalence of pathogens (8).

In the context of pathogens transmitted by wild birds, it is expected that in places where birds congregate in farms, the presence of many different species with diverse susceptibilities would make it more difficult for a pathogen to persist and spread, especially if a single species is the key reservoir for this pathogen. Meanwhile, the presence of species that are migratory on farms could increase the likelihood of the introduction of pathogens that these birds may have encountered on their migratory route. Finally, the risk of pathogen spillback from poultry to wild birds may also vary considerably with the species of wild bird encountering poultry or its feces.

Poultry farms attract wild birds due to water (puddles, canals, ditches) or food resources (spilled feed, drying feed, insects in manure, carcasses). These factors could increase the contacts between wild birds and bridge bird species, as well as increase the abundance of the latter and thus also contact between wild birds and domestic poultry (chickens, turkeys, game birds) (9). This contact can occur directly or indirectly through contamination of resources, thereby increasing the risk of transmission and spillback of avian pathogens, such as avian influenza viruses (AIV), Salmonella sp., and avian coronaviruses (10) among others. In this context, European starlings for example are a high-priority species for avian pathogen exposure detection studies as they can form large flocks in livestock feeders during the winter and autumn seasons, representing a potential risk of pathogen incursion into poultry farms, especially during the breeding season (11).

The unforeseen and unprecedented spread and change in the epidemiology of the highly pathogenic avian influenza virus (HPAIV) H5N1 of clade 2.3.4.4b, now fatally affecting new species, new continents, during all seasons, is decimating wild and domestic bird populations in much of the world, especially in the European and American continents (12). In contrast to other HPAIV it shows self-sustained prolonged transmission in wild birds and has already affected many poultry operations globally (12). This increases concerns regarding the potential transmission pathways of AIV by synanthropic bridge species. Migratory waterfowl, considered the main reservoirs for AIV (13, 14), play a key role in the introduction of many AIV subtypes through asymptomatic shedding, exerting a significant factor in the redistribution and transmission of these subtypes to domestic poultry (15). Several studies on the movements of migratory waterfowl have demonstrated their involvement in the large-scale spread of the virus (16). However, it should be noted that due to their ecological needs these wild birds rarely come into direct contact with poultry (17). In this scenario, synanthropic birds such as the house sparrow or the European starling are perceived as potential carriers and transmitters of AIV (18, 19), and could act as bridge both after exposure through direct contact with infected waterfowl, or contaminated environment in shared habitat (20).

Biosecurity protocols on farms rarely comprehensively assess how the virus enters the farm and which farm animals may be carriers of AIV. Despite some experimental evidence of the potential for synanthropic bird species to transmit AIV, there are very few studies dedicated to quantifying wild bird interactions with poultry farms. These studies include research in Australia using camera traps to monitor wild birds on different types of layer and meat chicken farms (21). Another study in the Netherlands quantified wild bird access to a free-range commercial laying hen farm by installing video cameras at a critical point for avian influenza (4). In southwestern France, a study used individual direct observations of wild birds on a free-range duck farm (5). Additionally, a recent study in northwestern Italy employed direct observations and camera traps on turkey and broiler duck farms, as well as laying hen farms (22). A study using satellite transmitter data from radio-marked waterfowl, showing occasional but regular incursions of marked birds onto poultry farm premises (23).

Collectively, these studies evaluate the accessibility of poultry farms for wild birds and identify the house sparrow as one of the most common species on farms due to its resident and sedentary nature. However, knowledge gaps exist regarding the frequency of farm visits by other species, seasonal changes in wild bird communities and characterization of sparrow interactions with other wild birds or even poultry on poultry farms.

The goal of this study was to generate data on seasonal changes of wild bird communities on different types of poultry farms and to investigate contacts of a key bridge species with other bird species that are non-residents on poultry premises and that could potentially lead to the acquisition and transmission of pathogens on to poultry. The latter is based upon the fact that in initial visits we observed a significant presence of house sparrows on poultry farms, commonly sighting them in barns and surrounding crop areas, even entering the barns where layers were housed/flight cages of red-legged partridges. Considering the persistence of AIV in bird feces and the environment (20) it has been confirmed that, under favorable conditions of high humidity and low temperature, AIV can persist in feces for extended periods, even in dry manure (24, 25). We hypothesize that in environments where house sparrows and other birds share resources such as food, water, or resting areas, contact could occur through shared surfaces contaminated by feces. If this occurs the house sparrow could become a potential vector for AIV as well as other pathogens.

For our purpose, we conducted camera trapping on the premises of various commercial layer and red-legged partridge farms in the Castilla-La Mancha region, in south central Spain at different time-points throughout the year corresponding to phenological events in wild bird ecology such as the breeding and wintering season as well as the periods during which migratory species conduct their spring and fall migration. We characterized the wild bird communities observed and used the house sparrow, which is the most abundant resident species, and the species most likely to also enter the layer barns/enclosures/flight cages as a potential bridge species. Hence, we analyzed the direct and indirect contacts of the house sparrow with other wild bird species observed in the camera traps. The collected data were used to quantify wild bird visits and their interactions with the house sparrow.

A total of 139,246 images were captured with the camera traps in the years 2018–2019. Among these, 78,779 images were taken on commercial layer farms, 30,187 on free-range layer farms, and 29,280 on red-legged partridge farms. A total of 31,816 birds belonging to 33 species, 21 families, and seven different orders were observed on the five farms. Out of these, 18 were resident bird species (55%), 11 were partially migratory birds (33%), and three were migratory bird species (12%). Most resident species belonged to the order Passeriformes (72%), such as the house sparrow, tree sparrow, and spotless starling (Sturnus unicolor) (see Supplementary Table S3).

The non-parametric estimators calculated 96.4% ± 3.6 and 95.2% ± 4.8% of the total observed species richness on Farms D and E, respectively. However, the non-parametric estimators suggest that species richness is higher on the remaining farms: Farm A (83.8% ± 16.2%), Farm B (69.2% ± 30.8%), and Farm C (55.4% ± 44.6%) (see Supplementary Figure S1, Supplementary Table S2).

We detected a significantly higher frequency of visits by wild birds on red-legged partridge farms as compared to caged layer farms and free-range layer farms (see Table 2, Figure 2). Visits detected on caged layer farms were less numerous, but from a much larger variety of species (Supplementary Table S3, Supplementary Figure S3). Species in the Columbiformes and Passeriformes order were significantly more likely to be detected (Table 2). Also, the number of bird visits detected was significantly higher during the breeding season (see Table 2, Supplementary Figure S3).

Resident bird species visited farms significantly more than partially migratory or migratory species (Table 2). Additionally, resident species had significantly more direct and indirect contacts with house sparrows as compared to migratory species (Table 3).

The house sparrow was the most frequently captured species in photographs, on all five farms and during all phenological periods. Spotless starlings were observed on four farms, being more abundant in the red-legged partridge farms, but not on the free-range layer farm (Supplementary Figure S4). In three of the farms, the camera traps recorded species such as the white wagtail (Motacilla alba), crested lark (Galerida cristata), and rock pigeon (Columba livia) (Supplementary Table S3). On the free-range layer farm, the house sparrow was less common (11%), and the Eurasian magpie (Pica pica) was the most frequently observed species (53%) (Supplementary Figure S4). Notably although anecdotical, waterbirds (mallard Anas platyrhinchos, black winged stilt Himantopus himantopus, ring-necked plover Charadrius hiatus) were detected on at least three of the farms (Supplementary Table S3).

Direct contacts between house sparrows and other species were significantly more likely during the breeding season and with resident bird species (see Table 3, Figure 2). The need to seek food and water, especially in juvenile birds, increases interactions with other species, particularly with the house sparrow. Indirect contacts with house sparrows (use of the same location by house sparrows within a time span of 24 h and thus potential of exposure to fecal contamination) was significantly more likely during the breeding season and least likely between house sparrows and Anseriformes and on the free-ranger layer farm (Table 4). Species that interacted more frequently with the house sparrow belonged to the order Passeriformes, being almost twice as common as the second most common type, the Columbiformes. Birds in the Charadriiformes order had fewer contacts with the house sparrow (see Supplementary Figure S4).

Our study applies camera trapping technology to the poultry farm environment to comparatively describe the composition and seasonal changes of farm bird communities on different types of layer/gamebird farms. This is to the best of the authors knowledge the first time such a study is carried out in Spain. Previous work and data collected in the present study provide evidence of the house sparrow as key species that enters layer buildings and flight cages of partridges and other bird species which has led to its designation, together with several other synanthropic bird species as potential bridge hosts (6, 9, 35) (Supplementary Figure S5). However, although bidirectional exchange of pathogens at the interface has been demonstrated and the potential of bridge hosts is generally accepted (36) little information exists yet on the frequency of contact and potential of contamination of resident farm birds by visiting migratory birds. For this reason we used the pictures obtained to also study the contact of wild birds that visit farm premises, but are unlikely to enter the buildings and enclosures, with the house sparrow (4, 5, 21).

The farms studied here are not directly connected to any wetland, which makes them theoretically unattractive to wild waterfowl (13), however satellite telemetry data has recently shown that waterfowl occasionally does make incursions onto poultry farm premises (23) and in fact our data shows, that even the studied farm premises are occasionally visited by waterbirds, either at open water tanks or temporary pools after heavy rains (Supplementary Figure S5).

The camera traps used for this study covered locations at the external fencing of the farms, aggregation hotspots such as water and food sources and possible entrances to the farm buildings. Actual farm bird diversity is certainly greater than that observed in this study. Camera traps do not always capture the total number of bird species visiting the farm, as they are positioned and focused on specific points, making it challenging to obtain a complete picture of the bird population. Additionally, even with increased sampling effort, as observed in farms A and B (n = 349 days; n = 92 days), we notice that the observed richness does not align with the studied estimators, indicating the need to extend our sampling period (Supplementary Table S2, Supplementary Figure S1). Logistical issues such as farm size, the number of cameras used, or limitations in data storage due to an abundance of individuals in a single location hinder obtaining the actual number of species. This is due to the size of the poultry farms, surrounding vegetation and habitats (various crops, buildings) (24) and changes in these (crop harvest, sowing etc.). However as the camera trap distribution design was similar for all studied farms and data was corrected for trapping effort, we can at least to some extend compare the collected information (21). Our results show that red-legged partridge farms attracted the highest number of visits, followed by caged and free-range layer farms. Most of the observed species were passerines. Possible reasons why wild birds were most attracted to partridge farms are the relatively easy access to food and water as both breeders and juvenile partridges are raised outdoors in batteries of breeding cages or large flight cages, respectively. Previous studies have shown that wild birds are not particularly attracted to free-range chicken farms, potentially as the grazing areas are rapidly degraded by the chickens while the large numbers of chickens also appear to intimidate most wild birds (24).

Among the species identified, the house sparrow is the most frequently observed and interacts directly and indirectly with a large variety of other species and a considerable number of individuals from which they could acquire pathogens including AIV as sparrows have been shown to be susceptible to AIV infection (18, 37, 38). The adaptation of the sparrow to human modified habitats, gregarious behavior and obligate commensalism drives their potential as bridge host (39). Other frequently detected species included the spotless starling and the rock dove. Both species are known to be frequently exposed to and carriers of pathogens such as Salmonella spp. and antibiotic resistance mechanism carrying Escherichia coli among others (10). Also, European starlings, a species closely related to the spotless starling have been experimentally shown to be able to transmit avian influenza virus to poultry (40). Species detected in free-range layer farms such as the white wagtail are consistent with the species detected in a study of free-living birds in enclosures of duck farms in France (5), while other species observed on the duck farms, such as cattle egrets, were observed less frequently.

Direct but also indirect contacts are often the main risk factor in pathogen transmission between wild and domestic birds (41). Contacts of other birds with house sparrows were observed generally in association to food and/or water and less frequently roosting space. Our results show that direct and indirect contacts of sparrows with other bird species on farms occur significantly more frequently with resident species than with partially or fully migratory species (Figure 3). If, in addition to direct contacts, the possibility of indirect contamination through secreta and feces is considered, with a maximum residence time of approximately 24 h, the potential for contamination of a sparrow by AIV or other pathogens doubles (Figure 3).

The highest number of direct and indirect contacts were recorded during the breeding season, followed by the spring and fall migration periods. For the wintering season, hardly any direct and indirect contacts were recorded even though the highest number of visits occurred at this time of year. As our farms are situated in a Mediterranean continental climate where food and water in natural habitats are more restricted in summer than in winter, likely during this period the availability of food and water was less important than other functions of the farm environment. More frequent farm visits during the breeding season, may be linked to the high number of juvenile individuals that rely on easily accessible resources. The high number of juvenile individuals increases the number of contacts with likely a higher number of naïve, more susceptible individuals thus increasing the probability of pathogen transmission (42). Also, during the breeding and post-breeding periods, the food requirements of breeding adult sparrows are at the highest, while it is under the Mediterranean continental climate the period with less food and water resources increasing the attraction to farm premises considerably. The lower number of direct and indirect contacts during fall migration is probably due to the abundance of food (cereal and fruits) in the season.

Both migratory and resident birds can be carriers of pathogens either directly or by exposure in contaminated environments and by Borie et al. (43) by contaminating resources, such as water or feed, with their droppings. Here we have detected few migratory or partially migratory birds on farm visits which in turn underlines the potential bridge host role of sparrows (18). While contact restriction measures are generally focused on the protection of poultry they also need to take into account the risk of environmental transmission from poultry to wildlife, by sewage, feathers, dust and aerosols from that can represent a major source of contamination for synanthropic wild birds such as the house sparrow that could than contaminate non-resident visiting birds during direct or indirect contacts (44).

Camera trap-based characterization allowed to describe composition and fluctuation of wild bird communities in different farm type environments across seasons and to identify species that could represent bridge hosts. Identifying the house sparrow as key species with resident populations on farms we evidence its connection to other bird species that visit the farm environment. Our results indicate that general biosecurity measures such as restriction o access to food and water are highly effective as the number of bird visits on red-legged partridge farms were significantly more frequent, than on other farms.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

AS-C: Formal analysis, Methodology, Visualization, Writing – original draft, Writing – review & editing. M-CC: Methodology, Resources, Writing – review & editing. YR: Methodology, Resources, Writing – review & editing. TC: Methodology, Resources, Writing – review & editing. UH: Conceptualization, Methodology, Resources, Writing – original draft, Writing – review & editing.