The spatial distribution of individuals within populations can result in fine-scale density-dependence and affect the social environment that is encountered. As such, it is important to quantify within-population spatial structuring and understand the factors that shape it. In this study, we make use of point process statistics to test whether colony size affects the statistical type of spatial nest distribution produced by common terns (Sterna hirundo) breeding at identical man-made rectangular and homogeneous islands of fixed physical size. Comparing sub-colonies of variable density both within and across years, we find that inter-nest distances are smaller at higher local and overall breeding density, but that the spatial distribution type does not vary across the observed densities. This suggests that the birds’ main settlement rules do not depend on density. In our case, analyses of fine-scale density-dependence or potential social effects therefore do not need to account for between-individual heterogeneity in settlement decision rules or acceptance of these rules. We urge, however, other studies to similarly test for density-dependence of the spatial distribution of individuals before undertaking such ‘down-stream’ analyses.

The common tern is a Holarctic colonially breeding and migratory seabird (Becker and Ludwigs 2004). The data we present here come from a long-term study population located in the Banter See at Wilhelmshaven on the German North Sea coast (53°36’N, 08°06’E). In 1992, 101 adult birds of this population were caught and marked with transponders (TROVAN ID 100; TROVAN, Köln, Germany), and since 1992 all locally hatched birds have similarly been marked with a transponder shortly prior to fledging.

The colony site consists of a line of six concrete islands (denoted A to F, land to lakeward; Becker 2015), each of which measures 10.7 x 4.6 m, is homogeneously covered with gravel, and is surrounded by a 0.6 m wall. Despite the distance between adjacent islands only being 0.9 m, they can be considered functional sub-colonies (Dittmann et al. 2007, Becker 2015). Three-times-weekly checks of the six sub-colonies are used to mark each nest, to assess laying date and to record reproductive parameters. During incubation, which is shared between partners, antennae are placed around each nest for 1–2 days to identify breeding individuals, such that we can distinguish between first, replacement, as well as second clutches. Since 1993, all nest sites have been mapped at the end of each breeding season by measuring the distance from the wall of one long and one narrow side of each island, leading to x- and y-coordinates of nest midpoints.

For this study, we include all first clutches for 2 out of 6 sub-colonies (C and F) and for 6 out of 26 available years (1996, 2000, 2005, 2010, 2015 and 2018; table 1). For each sub-colony and year, and for the moment when the last of the clutches was established, we interpret the midpoints of these clutches as a point pattern (figure 1) and use point process statistics (Illian et al. 2008, Wiegand and Moloney, 2014, Baddeley et al. 2016) to analyse this pattern.