Territory selection and cuckoo brood parasitism in Great Reed Warbler. Zlatia Bulgaria 2009

Study Extent

Daily data sampling within the field site.

Sampling

Materials and methods (a) Study area and study site The data was collected in June 2008 and April-June 2009, in a marsh near the village Zlatia (43°46’N, 23°30’E) situated in north-western Bulgaria. The marsh was chosen because of the relatively high parasitism rate of around 30% annually, exceeding 50% in some years (A. Antonov, unpublished data). The study site was defined by the marsh and surrounding vegetation and constitutes of a total area of approximately 2 km 2 . The marsh consists mainly of reeds, and some bulrush Typha latifolia, and it is surrounded by black poplar Populus nigra, white poplar Populus alba, cherry plums Prunus cerasifera, common walnut Juglans regia and acacia trees Acacia spp. Earlier, more than 20 years ago, the area was drained by a dike and used for agricultural production of corn and grass, in addition to grazing sheep. After the agricultural activity ceased the area has developed into the current marsh landscape. The dike made for drainage separates the fields and the high trees and is now due to eutrophication overgrown with reeds which widely shape the present habitat. The high and thick reeds are most abundant in parts of the area with adequate levels of water and are therefore located in the dike going abreast the high trees (see below). (b) Field procedures The first season, June 2008, was used to carry out preliminary field work, such as collecting morphological data and marking 58 great reed warblers with metal and colour rings for individual identification. Thus, nests were searched systematically in 2009 only. In June 2008 and from mid April to May 2009 great reed warblers were caught in mist nets and individually marked with metal and colour rings, and morphological data was measured as described by Svensson (1992). Catching of great reed warblers were carried out evenly throughout the whole marsh. Wing length (maximum chord to nearest 0.5mm), tarsus length (to nearest 0.1mm) and body mass (to nearest 0.1g) were measured from each territory holder and its female mate(s) if possible. Ageing of some males (n = 9) in 2009 was possible due to ringing data from 2008 and the sex was determined by the presence of breeding patch at first, then followed by a close monitoring of the colour ringed individuals throughout the breeding season. An index of condition was acquired by comparing the residual body mass in relation to wing length as the best morphological predictor of size (Gosler et al., 1998). After ringing the individuals, identification of males from the different territories were possible, in addition to determination of which males that were polygynous. A territory was considered occupied at the moment a singing male was observed. The territory holder and its mate(s) could later be confirmed when both parents were observed together by the nest. Territories were ranked by the order they became occupied by males, where the first were considered preferred and those occupied subsequently later were considered less preferred. The phenology of reed growth was not equal throughout the marsh and made some locations suitable at different stage, which was adjusted for by measuring reed at the same time of breeding in all territories. All nest sites were considered equally available and not constrained by nest site conflicts (Jones, 2001). Measurements were focused on the nest site exclusively, because of the importance of nest site characteristics in relation to brood parasitism and nest predation. Nests were systematically searched from early May until late June. Each nest was plotted on a GPS and further entered into Map Source, which made it possible to easily measure distance to surrounding territory variables e.g. trees, conspecific neighbours and reed bed edges. Nests of re-nesting birds in the territories were excluded from the data set. The initiation date of egg laying was recorded for each nest, and at mid-incubation, measures of vegetation structure were performed. To reduce disturbance during egg laying and early incubation, measures of vegetation structure were delayed to one week after initiation of incubation. Nest site vegetation structure was measured one square metre around the nest, which was considered to be the main area concerning the nest site. Reed thickness is traditionally used as a measure of reed quality (Dyrcz, 1981, Batáry & Báldi, 2005) and was also utilized in this study. I recorded it by using a digital calliper measuring the diameter of the thickest reed stem attached to the nest (Dyrcz, 1981). Several measures of reed structure were considered superfluous since they correlate to some extent (Ritterbusch, 2007), where the thickness is most important for the nest fundament. However, concealment of the nest is also influenced by the reed structure and was measured by the cover above the nest, measured by density of new reed shoots × mean height of new reed shoots above the nest. Reed density was assessed by averaging the number of touching new reed shoots at a 1m stick, sampled at three different locations randomly inside the square metre. Reed height was measured as the mean height of exclusively new reed shoots (measured to the nearest 0.1m above the nest). Old reed stems were excluded from the cover measure due to their little contribution in relation to cover compared to new reed shoots. Further, the average water depth beneath the nest was measured at 5 points (4 corners and centre of a square metre quadrate beneath the nest). Distance from the nest to the nearest cuckoo vantage point was measured from the nest and to the nearest tree above 5m (to the nearest 1m, using a GPS). Trees less than 5m were considered unfavourable as vantage points for cuckoos. Also, distance from the nest to the reed bed edge and distance from the nest to the closest active conspecific neighbour were measured (to the nearest 1m, using a GPS). The nests were visited each day, if possible, to check for parasitism and predation, and were considered successfully fledged if the nest survived until the nestlings were 8 days old. The nests were not further visited to minimize the disturbance of nestlings due to the risk of too early fledging.

Quality Control

Standard quality control i.e. summarized searches, distribution checks, etc.

Method steps

1. Mist net catching of adults, morphological measurements sampled, nest searched and recorded, nests and surrounding habitat measured, nestlings measured, and surrounding habitat characteristics measured.

The field site consists of a large swamp with a reed bed (habitat for Great Reed Warblers), surrounded by trees (mainly Populus sp.) which are utilized by Common Cuckoo as vantage points for parasitism.