A study of feral pigeon Columba livia var. in urban and suburban areas in the city of Jena, Germany

A study of feral pigeon Columba livia var. in urban and suburban areas in the city of Jena, Germany.— A population of feral pigeons, Columba livia var. was conducted in the city of Jena, Germany, from July to December 2007. Daily censuses were conducted by walking ten transects in a selected area of the city, five transects in built up areas and five in the suburbs. Pigeon population density was higher in urban areas than in suburbs but differences were not significant. Main behavioural activities recorded were resting, preening, flying, eating, sunning and roosting. Regular locations of activities were rooftops and roof edges in urban areas, and rooftops, eaves on balconies in suburban areas. The plumage phenotype most frequently recorded in both areas was Blue bar.

This present work is the first study of Columba livia var. in the city of Jena, Germany, and it reports on a daily census of behavioral activities, plumage phenotypes and breeding data at urban and suburban areas.

Materials and methods
The study was carried out from July to December 2007 in Jena city, Germany ( fig. 1) (50º 55' N, 11º 34' S). In an attempt to estimate the Columba livia population in the city, we chose a delimited area, including the built up centre and suburban areas. Five suburban east-west transects were defined and measured as followed: 1.28 km, 1.32 km, 0.46 km, 0.40 km, 0.39 km, with 0.25-0.35 km approximately between transects, occupying a total of 3.85 km 2 suburban area. Another five urban north-south transects were also measured: 1.02 km, 0.70 km, 0.80 km, 0.37 km, 0.40 km occupying a total area of 3.29 km 2 . Censuses were conducted using binoculars and digital camera four to five days per week. We recorded both stationary and moving birds in the transect dimension, two hours before and after midday following Uribe et al. (1984), Senar & Sol (1991), Montalti & Kopij (2001). Pigeon density (individuals per km 2 ) was calculated from the census transects per area. Urban areas were characterized by high buildings, churches, cafés and food shops, buses and cars. Such buildings resemble rocky sea cliffs, an ideal synantropic environment for feral pigeon settlement. Suburban areas consisted of residential houses, few shops and little traffic.
Behavioural activities were recorded using binocular observation on daily census while walking on street transects. We registered resting, preening, sunning, eating, flying, nesting, roosting, drinking, sleeping, perching, and mating.
Plumage phenotypes were recorded using binoculars, camera and bird guides on daily census from July to December. These data were later classified per month and per area (Goodwin, 1970;Murton et al., 1973Murton et al., , 1974Johnston & Janiga, 1995;Miller, 1997).
Breeding activity was measured monthly (seven visits) at the tower of Lobeda locality (50º 52' N, 11º 35' E), 7 km from Jena. The tower contained twenty experimental wooded nest boxes occupied by pigeons where we classified bird stages. At Jena, breeding was measured on transects by registering nesting on walking census in both areas.

Results
Feral pigeons density during census was higher in urban transects, 730 ind/km 2 ± 199 with a total number of individuals of 2,377 (61%), than in suburban transects, 392 ind/km 2 ± 205, with a total number of 1,510 (39%) pigeons. (fig. 2). However, differences between the two areas were not statistically significant (α = 0.05; F = 1.52; p > 0.24). A t-test, considering different variances analysis, showed no significant differences (p > 122; t > 1.23). Mean values of pigeons per transect were 396 in urban areas and 251 in suburban areas.
Columba livia behavioural activities were registered on transects in both areas. Comparison of pigeon activities showed no significant differences between urban and suburban areas in any month except December (α = 0.05; F = 7. 33) (table 1, fig. 3).
Breeding activities in both localities and areas were performed by checking presence of birds, nesting and mating (table 1) and registering number of adults N = 12 ± 2, fledglings N = 10 ± 1, small chicks N = 6 ± 1, eggs N = 13 ± 1, and number of active N = 25 ± 2 and non-active nests N = 13 ± 2. Variance analysis showed no significant differences between these bird stages at the tower ( fig. 5).

Discussion
Feral pigeon urban density was higher than suburban density but differences were not statistically significant. Urban areas were characterized by high buildings that were architectonically favourable for pigeon settlement: old, grey, abandoned, rocky-walled churches and some empty domes were favoured. In suburban areas they favoured greenish landscape, low houses. Differences in population density were expected between areas but lack of significant findings may be due to the boundaries chosen Jena may not correspond structurally to a classical European gothic city. The city was largely rebuilt after World War II, leaving a mix of old centre patches in the middle of suburban neighbourhoods. To find differences between areas it could be interesting to conduct a study comparing the outer city with a suburban area. Blue bar (62 %) was the most abundant plumage followed by Checker (19 %) and T Checker (10 %). The study of a probable phenotype-dependant selection in an urban environment such as Vienna showed that natural distribution was 30 % of Blue bar and 28 % of Blue Checker. Bird activities were similar in both areas without significant changes.