Gill Caradoc-Davies, of Portobello, asks :-

About 4 years ago, we noticed that the starlings that used to roost on Pudding Island changed their behaviour, and started to come in huge flocks to roost in the gum trees behind our house, making a very loud and cheerful chattering to brighten up our winter dawns and dusks.

We noticed that they only do this in large flocks from late autumn to early summer, and presumed that after this time they pair off to raise their chicks.

In the transition between these two phases, they still come in to roost in large numbers in the gums, but they come in PAIRS.

We keep a diary. This transition lasted for the last two weeks of October during which a few still come to the gums, but not so obviously in pairs. Is this typical behaviour of starlings?

Ian Jamieson, an ecologist at the University of Otago, responded.

Starlings were introduced between 1860 and 1880 in small numbers throughout New Zealand including Otago, and now are common and widespread across the country. Starling roosting behaviour has been well studied in North America and Europe, where they are known to gather in the hundreds of thousands in communal nocturnal roosts, especially over winter.

Communal roosts are believed to provide safe havens from predators at night, but there could be additional benefits such as following other birds in the morning to find new foraging sites. During spring, the large winter roosts break-up and smaller roosts are formed closer to nesting areas.

Starlings nest in loose colonies in tree cavities, banks or under eves of buildings and males in particular will vigorously defend their nest cavity all year round. Starlings forage alone or in smaller flocks during the spring and summer, but pairs will visit their nest site in early morning and evening before returning to communal roosts at twilight. One study in France found that birds from breeding colonies develop local dialects in their calls, and birds that share the same dialect will position themselves close together in a communal roost made up of many individuals from several breeding colonies. This could explain why starlings are so vocal as they settle into their roosts at night. Therefore it would not be surprising in spring to see pairs of birds arriving at the roost in the evening (as observed at the Portobello starling roost) and leaving together in the mornings to visit and defend their nest site before starting to forage for the day. However, I could not find a specific reference to the seasonal arrival of pairs at the roost, as described here. Both parents share incubation duties during the day and only the female incubates at night, but it is unknown whether the male returns to the communal roost at night, although he occasionally roost next to the incubating female.

The observations of roosting starlings at Portobello are interesting and nicely illustrate the change in behaviour across the seasons.