Effects of agri-environmental habitat provision on winter and breeding season abundance of farmland birds

•We monitored birds for five years across varying levels of agri-environmental uptake.•Agri-environmental treatments significantly increased winter bird numbers.•For many species this led to higher numbers in the breeding season.•Spatially separated controls were important for detecting treatment effects.•Agri-environment schemes can benefit farm-scale populations of many bird species. Farmland bird populations continue to show declines in spite of over 20 years of research and implementation of agri-environmental schemes (AES) intended to reverse this. Although it is well known that provision of winter food resources can attract farmland birds, there is continuing uncertainty over the ability of AES to provide tangible benefits for target species in terms of increased abundance. Answering these questions is hampered by interannual fluctuations in bird populations and the mobility and territoriality of farmland birds, which have complicated the interpretation of previous studies. We monitored birds for five years on a large arable estate in central England managed under varying levels of AES uptake (low level uptake of simple and widely applicable AES options, more extensive uptake of more complex AES options), and two control treatments (on-site and off-site). Bird abundance in winter and both total abundance and number of territories in the breeding season were calculated from monthly visits to 16 transects. Several species showed significantly higher winter abundance on AES treatments, particularly granivorous species (e.g. reed bunting, yellowhammer, linnet). Many other species (e.g. blackbird, chaffinch, robin) also showed significant differences in winter abundance between treatments on the estate and off-site controls. In the breeding season, linnet, reed bunting, goldfinch and combined granivorous birds showed higher abundance or number of territories on AES treatments compared to on-site controls. For most other species the differences were only significant between treatments on the estate and off-site controls. Independently of AES treatment, a lower coverage of cereals or greater Shannon diversity of crops in the local landscape also had a positive effect on the abundance of many species. Our results suggest that well-implemented AES can significantly enhance local populations of both farmland specialists of conservation concern and generalist species. Our results also show that, in many cases, these effects were only demonstrable at the farm sc