Our understanding of the impact of feeding wild birds is far from complete, but we are beginning to unravel the effects of providing foods at garden feeding stations. An important area of research has been to examine how supplementary foods shape populations through its impacts in individuals. Feeding wild birds is a popular pastime and many of us provide seed and other foods to help our feathered friends. But what impact does all this food have? It is a huge resource and one that can increase overwinter survival and bring forward the timing of breeding, but we also know that the feeding of wild birds has been linked to the transmission of disease. Research by Kate Plummer and colleagues provides new insight into one particular aspect of food provision – how it shapes bird populations. Kate’s research has examined the extent to which the provision of supplementary food during the winter months influences the physiological condition of individuals and populations the following breeding season. By using woodland populations of Blue Tits, Kate and fellow researchers have been able to compare the effects of providing fat, and fat plus vitamin E (an antioxidant), against a control population of unfed birds. The feeding carried out during the winter months ended at least a month before the tits began egg-laying. Provisioning with fat and vitamin E was found to alter the composition of Blue Tit populations, such that they included birds that had been in significantly poorer condition prior to feeding. Because those individuals were found to have lower levels of carotenoids in their breast feathers than unfed birds, Kate was able to conclude that supplementing with vitamin E and fat in winter had altered the survival and recruitment prospects of these lower quality individuals; lower levels of carotenoids are indicative of poorer physiological condition. However, provisioning with fat alone was found to have a detrimental impact on breeding birds. It appears that the provision of supplementary foods during the winter months can alter both the structure of the breeding population the following season, and the condition of individual breeding birds. Such effects may have consequences even longer term; through this work, for example, it was found that individuals with higher blood plasma concentrations of malondialdehyde (which is indicative of oxidative damage) produced offspring that were structurally smaller and which suffered from reduced fledging success. The importance of antioxidants, like vitamin E, can also be seen from Kate Plummer’s earlier work on yolk mass. While Plummer et al. (2013) found that winter provisioning with fat subsequently impaired an individual’s ability to acquire, assimilate and/or mobilise key resources for yolk formation, this was not the case where vitamin E was also included in the food presented. A high fat diet, such as that potentially obtained from the food provided at garden feeding stations, may well increase the requirement for antioxidants in order to combat the greater levels of oxidative damage associated with a diet rich in fats. Clearly, there is still much to learn about how the provision of supplementary food affects wild birds and their populations.

The increasing temperatures associated with a changing climate may disrupt ecological systems, including by affecting the timing of key events. If events within different trophic levels are affected in different ways then this can lead to what is known as phenological mismatch. But what is the evidence for trophic mismatch, and are there spatial or temporal patterns within the UK that might point to mismatch as a driver of regional declines in key insect-eating birds? A changing climate is leading to changes in the timing of key ecological events, including the timing of bud burst, the spring peak in leaf-eating caterpillar biomass and the timing of egg-laying in many bird species. If the timings of these different events shift at different rates then there is a danger that they may get out of synch with one another, something that is referred to as phenological mismatch. This may be a particular problem for birds like Blue Tit, Great Tit and Pied Flycatcher, which time their breeding attempts to exploit the spring peak in caterpillar abundance. Much of the recent work on mismatch and its impacts on the fitness and population trends of caterpillar-eating birds has looked at changes over time. However, it is also possible for mismatch to vary in space if species respond differently in different areas, perhaps because of local adaptation to geographic variation in the cues that they use. This paper looks at mismatch in both space and time, using information from three trophic levels, namely trees, caterpillars and caterpillar-eating birds. While information on bud burst came from 10,000 observations of oak first leafing for the period 1998-2016, that for caterpillar biomass was inferred from frass traps set beneath oak trees at sites across the UK for the period 2008-2016. Bird phenology data came from the ‘first egg date’ values calculated from 85,000 nest records of Blue Tit, Great Tit and Pied Flycatcher. The focus of the work was on the relationship between the phenologies of these interacting species; where timing changes more in one species than the other, this is indicative of spatial or temporal variation in the magnitude of mismatch. The results reveal that, for the average latitude (52.63°N) and year, there is a 27.6 day interval between the timing of oak first leaf and peak caterpillar biomass. With increasing latitude, the delay in oak leafing is significantly steeper than that of the caterpillar peak. At 56°N the predicted interval between these two trophic levels drops to 22 days. In the average year and at the average latitude, the first egg dates of Blue Tits and Great Tits were roughly a month earlier than peak caterpillar biomass, meaning that peak demand for hungry chicks occurred soon after the peak in resource availability. Interestingly, peak demand in Pied Flycatchers occurred nearly two weeks later than peak caterpillar availability, suggesting a substantial trophic mismatch between demand and availability for this species within the UK. However, it is worth noting that Pied Flycatchers provision their nestlings with fewer caterpillars and more winged invertebrates compared to the tit species studied, so they may be less dependent on the caterpillar peaks. The work also revealed that the timing of first egg date between years varied by less than the variation seen in timing of the caterpillar resource peak, which gave rise to year-to-year variation in the degree of mismatch. For every 10 day advance in the caterpillar peak, the corresponding advance in the three bird species is 5.0 days (Blue Tit), 5.3 days (Great Tit) and 3.4 days (Pied Flycatcher). In late springs, peak demand from the tits is expected to coincide with the peak resource availability, with flycatcher demand occurring shortly after. In early springs, the peak demand of nestlings of all three species falls substantially later than the peak, leaving the three mismatched. Warmer conditions also shortened the duration of caterpillar peaks. One of the key findings of the work is that in the average year there is little latitudinal variation in the degree of caterpillar-bird mismatch. This means that more negative declines in population trends of certain insectivorous birds in the southern UK, driven by productivity, are unlikely to have been driven by greater mismatch in the south than the north. The lack of evidence for latitudinal variation in mismatch between these bird species and their caterpillar prey suggests that mismatch is unlikely to be the driver of the spatially varying population trends found in these and related species within the UK.

Cuckoos have been documented using a range of host species but most of their eggs are laid in the nests of just a small number of species. Why do they select these species and to what extent might other species be suitable hosts, as yet unused to any great degree? The Cuckoo is a generalist avian brood parasite, known to have utilized at least 125 different bird species as a host within Europe. Despite this, individual female Cuckoos are thought to be host-specific, preferentially laying their eggs in one – or a few – host nests. This has led to individual female Cuckoos being classified into host races, termed ‘gentes’, some 20 of which have been described within Europe. Given the large number of songbird species breeding in Europe, and the relatively few female Cuckoo gentes, it raises the question of what makes a good host. By examining data from three different sources, including nest record data from the BTO, Bård Stokke and colleagues sought to establish which characteristics determined the suitability of a songbird species as a potential Cuckoo host. Some of the characteristics they investigated included nest placement, chick diet, habitat, abundance and the timing of the breeding period. The results of analysing two independent data sets, one from the UK and one from Germany, produced similar results. The analyses revealed that species breeding in forest and rocky areas were used less than species breeding in other habitats; that those feeding their nestlings with plant material were used less often than those feeding their nestlings with invertebrates (though not significantly so in the UK) and that hosts tended to be of intermediate body size and have larger populations. The analyses also showed that, perhaps unsurprisingly, species nesting in cavities are used less frequently than those nesting elsewhere. Nest height, nest depth and overlap in the breeding period did not appear to affect parasitism in any of the data sets examined. Through the work the researchers were able to calculate a ‘host suitability index’ for each of the European species in the study. Interestingly, there was no evidence of a clear separation between suitable and unsuitable host species’ instead, there appeared to be a continuum from low to high suitability. However, all of the species with a recognized corresponding Cuckoo gens are ranked towards the high suitability end of the continuum. The host suitability index could indicate species that might become the focus of Cuckoo nest parasitism in future if, for example, the host population increases because of climate- or habitat- change.

Some bird species provide cultural services, being aesthetically pleasing and having behaviours that people find interesting to watch. Others provide disservices (e.g. gulls, pigeons and corvids) negative for well-being. By documenting how the abundance and richness of species in these two groups correlates with human population density it was apparent that socio-economically deprived areas support low ratios of birds to people, particularly of cultural service species. These results inform management of green space, and provision of feeding and nesting sites, to promote positive interactions between birds and people within urbanised landscapes. Working in collaboration with the University of Exeter, and funded by NERC, researchers carried out extensive bird surveys within an urban area, centred on the towns of Milton Keynes, Luton and Bedford, as part of a wider project investigating urban ecosystem services. These provided measures of the abundance and richness of bird species within both the cultural services (35 species) and disservices (9 species) groups. The research team was able to look at the human population by using data from the 2011 National Census, and to assess socio-economic status by using information published by the Office of National Statistics. Since bird diversity is strongly associated with the structure and availability of urban green space, the team also had to factor in the green space present within the study area. Analyses revealed that the abundance of cultural service species increased with human population density but peaked at c.1,100 people per 500mx500m grid tile. The abundance also increased with the proportion of urban green space. Interestingly, the species richness of cultural service birds decreased with human population density but increased with percentage green space. There was a positive linear relationship between the abundance and richness of cultural disservice species and both human population density and the availability of green space. When the researchers mapped how the abundance of service and disservice birds co-varied with human population density, they found that the two groups of birds showed distinctly different spatial patterns. Service species were most abundant in areas of medium housing density – the suburbs – while disservice birds were most abundant in areas of dense housing, such as those around urban centres. While these different patterns are not a direct consequence of human population density per se, they probably result from spatial differences in urban form, the pattern and management of urban green space, levels of disturbance and the availability of resources, all of which are known to vary along socio-economic gradients. This underlines that people living in different parts of the urban landscape are likely to experience different relationships with wild birds, with the human communities in socially deprived areas exposed to more species with negative behaviours than wealthier communities. A consequence of this is that the increased frequency of negative interactions experienced by these people is likely to shape their connection with nature and support for the conservation of the natural world in a negative manner. The study identifies opportunities to deliver management approaches to counter these unfavourable relationships. Investment in urban green space and its management for cultural service birds is one obvious option, but there are also opportunities at the householder level, through practices such as wildlife gardening. Such householder level approaches can be of wider benefit because their beneficial effects are likely to increase the abundance and richness of cultural service birds in neighbouring gardens, meaning that the actions of a small number of people can provide health benefits for the wider community.

Work on emerging infectious diseases and garden birds in the UK has been supported by citizen science projects, most notably Garden BirdWatch, Garden Wildlife Health and the Garden Bird Health Initiative – the latter now superseded by Garden Wildlife Health. Through these schemes, researchers have been able to carry out national surveillance of emerging diseases, including finch trichomonosis, Paridae pox and passerine salmonellosis. This paper, part of a special issue of Philosophical Transactions focusing on wildlife disease issues, reviews the work that has been carried out on these diseases over the past 25 years. It also takes a look at the occurrence of mycotoxin contamination of food residues in bird feeders, which also pose a risk to the health of wild birds. A citizen science approach provides a cost-effective means to undertake large-scale and year-round disease surveillance (Garden Wildlife Health), delivered in parallel to the monitoring of wildlife populations (Garden BirdWatch). By combining large-scale surveillance and targeted post-mortem examinations we can differentiate between the multiple diseases that result in non-specific clinical signs (e.g. lethargy and a fluffed-up appearance). The sample archive collected through post-mortem examination also enables future identification of other disease agents, including those – such as environmental pollutants – associated with non-infectious disease. Avian trichomonosis, caused by the protozoan parasite Trichomonas gallinae, has long been known to affect pigeons, doves and birds of prey. Its emergence in finches in 2005 led to a major population decline in Greenfinch and Chaffinch, reported in an earlier paper. While multiple strains of Trichomonas gallinae are known to infect pigeons and doves in the UK, a single clonal strain is responsible for the epidemic seen in finches. Quite why Greenfinch is so susceptible to the disease is unclear. It is likely that the disease spilled over into finches following the increase in use of garden feeding stations by Woodpigeon, a recent change that follows an increase in wider countryside populations. Avian poxvirus has been documented in a number of garden bird species and is most often seen in House Sparrow, Starling, Woodpigeon and Dunnock. Its emergence in UK tits saw a more severe form of the disease, resulting in pronounced skin lesions, some of which were likely to have hampered the individual’s ability to feed and to avoid predators. Sequence analysis of the poxvirus strains affecting garden birds revealed that a single clade is responsible for the disease seen in UK tits. This form has been known in Scandinavia since the 1950s, with incidents seen elsewhere in mainland Europe since 2005. Because UK tits are relatively sedentary in their habits, and because of the geographical pattern of disease spread seen, it is likely that the disease reached the UK via a biting insect – such as a mosquito, crossing the English Channel in a warm plume of air. Salmonellosis has been reported in wild birds since at least the 1950s, with the bacterium responsible known to be capable of persisting in the environment for many months. Greenfinch and House Sparrow are the two species in which the disease is most often seen. Passerine salmonellosis incidents have a clear seasonality, peaking in January. Interestingly, the prevalence of the disease in UK passerines has dropped sharply over recent years. This may reflect increased immunity to the particular form (DT56v) that had been seen here; or it may be that transmission is density-dependent, with the sharp decline in Greenfinch populations resulting in much lower rates of transmission. Mycotoxins, which include the aflatoxins and ochratoxin, are secondary metabolites produced by certain fungi of the genera Aspergillus and Penicillium. Exposure to the aflatoxins and ochratoxin can exert a range of adverse effects in birds, and the fungi involved and their toxins are known to occur on foodstuffs, including peanuts. Food residues from bird feeders were screened for the toxins as part of the current study, with detectable aflatoxin residues found in all seven samples, two of which greatly exceeded the maximum permitted limits set for such residues in peanuts destined for livestock feed, which includes wild bird food. It therefore seems likely that garden birds may be exposed to these toxins at levels associated with toxic effects in captive birds. This review underlines the great deal of new information that has been generated through these citizen science projects. It also highlights future research needs, particularly around the identification of risk factors, and that we need to understand the balance of risks and opportunities that garden bird feeding provides.