Habitat quality, individual variation and dispersal in the great tit: population consequences

Most organisms live in environments that vary, with good and poor quality patches interspersed. Most organisms also disperse at some stage in their lives: usually the major dispersal event comes between birth and when an animal settles and begins to reproduce. Traditionally, we think of dispersal as a process that counter-acts the evolution of differences between, and within, populations. As individuals move around their habitat, they effectively re-shuffle the characteristics of the population during each consecutive generation. However, the fact that the environment varies means that some places are better spaces to settle than others, and that there should be competition to settle in these places. If the ability to win this competition is predicted by an individual's characteristics (for example, its size, or time of birth), then dispersal can actually be the cause of differences between, or within, populations. Rather than re-shuffling populations, non-random dispersal of this kind can actually act to sort individuals. Some recent work on a common British bird, the great tit, studied near Oxford, suggests that this is exactly what is occurring. Over time, the larger individuals in the population tended to accumulate in the better quality habitats, whereas the smaller individuals tended to accumulate in the poorer habitats. This produced an evolutionary change in the sizes of the birds in different parts of a single continuous population. The aim of this work, is to test various components of the idea that dispersal can be a non-random process, depending on the characteristics of the individual concerned, and that this non-randomness depends on how variable the environment is. We will study a population of great tits near Oxford, that has been studied since 1947, and for which a great deal of information is already available regarding the quality of the environment, and the way that individuals disperse (for example, over 4000 individual nestlings have been tracked from their place of birth to their eventual breeding location). However, despite the wealth of data, we know remarkably little about the actual processes involved in dispersal, in terms of how individuals use space between their birth site and their breeding site. Our first aim will be to collect dispersal data, using a new, automated, tracking system, where birds are registered and identified as they visit artificial feeding sites placed around the woodland. Our second aim is to develop a better understanding of the key factors that determine the environmental quality of different sites for the breeding birds. We will focus particularly on the breeding density of individuals, and the timing of leaf development of oak trees near to nest sites; preliminary data suggest that oak trees are a very important resource for these birds, probably because the main food supply for their offspring is caterpillars that feed on oak leaves. Our third aim is to carry out an experiment that manipulates the quality of the environment, to see how this affects the characteristics of the birds that settle and breed there. We will do this by creating different densities of potential nest-sites. We know that great tits will settle at artificially high densities, if nest-sites (boxes) are provided in great abundance. The birds breed less successfully at high density because of competition for resources, meaning that these sites should be less attractive than when a population is at low density, i.e. when nest-sites are in short supply. Finally, we will carry out manipulations of the characteristics of individual birds, to test how this affects the quality of the sites these individuals settle in. This is most easily done by moving a few young between pairs of nests: the birds that are reared in reduced broods grow better and should therefore be more competitive, and able to disperse to better sites than those that have grown up in enlarged broods.