Nocturnal flight calling behaviour of thrushes in relation to artificial light at night

Migratory birds are subject to many pressures during their life cycle and many are declining as a consequence. Evidence from North America shows that for species that migrate at night, bright artificial light sources associated with urban areas can disrupt natural movement patterns, leading to direct and indirect fitness consequences. Comparable evidence for species and urban areas in Europe is limited. This study aimed to measure the response of nocturnally migrating thrushes to artificial light at night in the UK. We used passive acoustic recorders deployed across a gradient of artificial lighting to record the flight calls of three thrush species, with an expectation of greater call rates over brightly lit areas. We trained a convolutional neural network automatically to locate and identify thrush calls in the audio recordings, achieving area under the curve (AUC) values in withheld validation data of 0.93-0.98, and recall on independent field data of 85-94%, depending on species. Seasonal patterns of call rates were positively correlated across sites but there were large differences in absolute rates between sites. Call rates were up to five times higher over the brightest urban areas compared with darker villages, suggesting a strong phototaxic effect of artificial light at night on migratory thrushes. These results confirm that monitoring of flight calls can provide valuable information on the timing of nocturnal migration, but that the effects of artificial lighting must be taken into account in any comparisons of abundance across sites. European cities are not blighted by mass mortality of migrants striking illuminated buildings; even so, these results show that nocturnal migrants are influenced by light pollution. Ascertaining whether this has fitness consequences is a priority so as to inform the design and illumination of future urban areas.