Increased adult mortality and reduced breeding success with age in a population of common guillemot Uria aalge using marked birds of unknown age

Although there is general consensus about the existence of senescence in vertebrates, empirical evidence of senescence in demographic parameters in wild populations is limited. Data on breeding success and survival of breeding common guillemots Uria aalge were collected over 20 years on the Isle of May (Scotland) using a pool of individuals marked as adults. Because the years of hatching of individuals were not known, we used the time (years) elapsed since first capture (TFC) as a measure of age. The use of this proxy did not create any bias in estimating senescence in the case of a linear decline, nor did it greatly decrease the power of a test for senescence. Breeding success declined significantly from 0.81 (95% CI: 0.77–0.84) to 0.62 (0.54–0.68) over the study period. It also varied in relation to age, initially increasing (from 0.62 (0.54–0.68) at TFC of 0 year to 0.76 (0.73–0.79) at TFC of 9 years) up to a plateau (from TFC of 9 years with 0.76 (0.73–0.79) until TFC of 13 years with 0.77 (0.74–0.79) before declining in later life to 0.70 (0.61–0.78) at TFC of 19 years). Survival was generally high and varied significantly from year to year. It also declined with TFC: survival of birds marked in 1982 decreased from 0.92 (0.85–0.96) at a TFC of 0 year to 0.88 (0.82–0.92) at a TFC of 19 years. Resighting probabilities also declined with TFC suggesting that the oldest birds do not come back to the colony to breed as regularly as younger individuals. These findings indicate that individual common guillemots on the Isle of May showed both actuarial and reproductive senescence.     

The incidence of the tick parasite Ixodes uriae in Kittiwake Rissa tridactyla colonies in relation to the age of the colony, and a mechanism of infecting new colonies

Kittiwakes Rissa tridactyla (683 chicks and 64 adults) were examined for ectoparasites in three breeding areas: Isle of May (Firth of Forth, Scotland; eight colonies of approximately known age of occupation on natural sites), North Shields and Gateshead (Tyne and Wear, England; well established colonies of known age on buildings). All the 208 ticks collected were Ixodes uriae. On the Isle of May, a colony which had only been established for 7 years was already infested by ticks. Tick incidence increased significantly with the length of colonization of the area during the first 30 years. Age of colonization did not have an effect after 30 years. In Gateshead only a single adult among the 127 birds examined had a tick. No ticks were found on the 106 chicks born at North Shields checked in 1987 or in 1988. However, among nine visiting newly-fledged chicks from neighbouring colonies caught at the North Shields colony, six had either ticks or scars of previous infestation. Some visiting chicks remained for several days in the host colony and were even fed by the owners of visited nests. Ixodes uriae may thus spread from one colony to another during visits of newly-fledged infested chicks to neighbouring colonies. The absence of any infestation in old colonies (38 years in North Shields and 17 years in Gateshead) on buildings that are regularly visited by newly fledged infested chicks, suggests that buildings may lack the rock refuges necessary for tick survival from one year to the next. In this respect Kittiwake colonies on buildings are rather atypical.     

Intercolony movement and behaviour of immature Guillemots Uria aalge

A minimum of 61 Guillemots Uria aalge ringed as chicks at other colonies was recorded at the Isle of May between 1987 and 1991 (four prior to 1990, six in 1990, 51 in 1991). Two were adults and 59 were immatures. Most of the British and Irish colonies where many chicks had been ringed were represented. Higher proportions of birds ringed at colonies relatively close to the Isle of May were observed compared to birds ringed at more distant sites. Visitors ranged in age from 2 to 5 years, with single 10- and 14-year-olds. Relatively fewer 2- and 5- than 3- and 4-year-old birds were seen. Visiting birds were usually recorded once only, significantly less often than native immatures of the same age, except for 2-year-olds. The proportion of birds occupying sites on intertidal rocks as against sites in the colony appeared to be higher compared with native birds of the same age. Visiting birds formed a substantial proportion of the immature population. Ten immature birds from the Isle of May were recorded elsewhere. Four of these had been seen on the Isle of May previously, and four were seen there subsequently. A 6-year-old hatched at Sumburgh ( c. 420 km by sea) bred at the Isle of May in 1992.     

Investigating the effects of age‐related spatial structuring on the transmission of a tick‐borne virus in a colonially breeding host

Higher pathogen and parasite transmission is considered a universal cost of colonial breeding due to the physical proximity of colony members. However, this has rarely been tested in natural colonies, which are structured entities, whose members interact with a subset of individuals and differ in their infection histories. We use a population of common guillemots, Uria aalge, infected by a tick‐borne virus, Great Island virus, to explore how age‐related spatial structuring can influence the infection costs borne by different members of a breeding colony. Previous work has shown that the per‐susceptible risk of infection (force of infection) is different for prebreeding (immature) and breeding (adult) guillemots which occupy different areas of the colony. We developed a mathematical model which showed that this difference in infection risk can only be maintained if mixing between these age groups is low. To estimate mixing between age groups, we recorded the movements of 63 individually recognizable, prebreeding guillemots in four different parts of a major colony in the North Sea during the breeding season. Prebreeding guillemots infrequently entered breeding areas (in only 26% of watches), though with marked differences in frequency of entry among individuals and more entries toward the end of the breeding season. Once entered, the proportion of time spent in breeding areas by prebreeding guillemots also varied between different parts of the colony. Our data and model predictions indicate low levels of age‐group mixing, limiting exposure of breeding guillemots to infection. However, they also suggest that prebreeding guillemots have the potential to play an important role in driving infection dynamics. This highlights the sensitivity of breeding colonies to changes in the behavior of their members—a subject of particular importance in the context of global environmental change.     

Within- and between-year variation in the juvenile survival of Common Guillemots Uria aalge

We studied juvenile survival of 20 cohorts of Common Guillemot Uria aalge chicks colour-ringed on the Isle of May, Scotland, using both live observations at the colony and dead recoveries, allowing estimation of fidelity to the colony as well as survival. In this seabird, chicks leave the colony when only partly grown and are cared for by the male parent for several weeks afterwards. First-year survival varied strongly between cohorts, with a mean of 56% (range 30–91%). We did not identify any covariates which could explain this variation, whether relating to climate, population size or prey density. Survival was low during two regime shift episodes in the North Sea (1987–90 and 2000 onwards). Early hatched chicks were substantially more likely to survive than those hatching later in most years, whereas body condition at ringing had no detectable effect. Ringing recoveries indicated that mortality was highest in mid-winter, i.e. well after the cessation of paternal care. These results do not support the hypothesis that variation in prey quantity or energy content before fledging is a primary driver of variation in juvenile survival. Rather, it seems that chicks of high-quality parents are more likely to survive, as high-quality females tend to lay earlier in the season, and high-quality males presumably are better able to prepare their chicks to survive their first winter at sea. Very few (4%) Guillemots emigrated permanently before age 3 years, but from age 5 onwards 25–30% of birds annually left the colony or otherwise became unobservable.     

Survival of adult Common Guillemots Uria aalge at three Scottish colonies

We present the results of a mark–recapture analysis of survival rates of Common Guillemots with at least one year's breeding experience for birds from three Scottish colonies. Estimates of overall survival of adults from the Isle of May (North Sea), Canna and Colonsay (both off northwest Scotland) from an analysis combining data from the three colonies were 94.8% (se = ±0.6), 92.4% (se = ±0.9) and 96.7% (se = ±0.6) respectively. The rates from Canna (but not Colonsay) differed significantly from those of guillemots on the Isle of May. The Canna and Colonsay figures are the first survival estimates for guillemots in northwest Scotland, an area of high conservation concern for this species. Low input, long-term studies of seabirds can produce important data on adult survival.     

Variance in prey abundance influences time budgets of breeding seabirds: evidence from pigeon guillemots Cepphus columba

We use data on pigeon guillemots Cepphus columba to test the hypothesis that discretionary time in breeding seabirds is correlated with variance in prey abundance. We measured the amount of time that guillemots spent at the colony before delivering fish to chicks ("resting time") in relation to fish abundance as measured by beach seines and bottom trawls. Radio telemetry showed that resting time was inversely correlated with time spent diving for fish during foraging trips (r=−0.95). Pigeon guillemots fed their chicks either Pacific sand lance Ammodytes hexapterus , a schooling midwater fish, which exhibited high interannual variance in abundance (CV=181%), or a variety of non-schooling demersal fishes, which were less variable in abundance (average CV=111%). Average resting times were 46% higher at colonies where schooling prey dominated the diet. Individuals at these colonies reduced resting times 32% during years of low food abundance, but did not reduce meal delivery rates. In contrast, individuals feeding on non-schooling fishes did not reduce resting times during low food years, but did reduce meal delivery rates by 27%. Interannual variance in resting times was greater for the schooling group than for the non-schooling group. We conclude from these differences that time allocation in pigeon guillemots is more flexible when variable schooling prey dominate diets. Resting times were also 27% lower for individuals feeding two-chick rather than one-chick broods. The combined effects of diet and brood size on adult time budgets may help to explain higher rates of brood reduction for pigeon guillemot chicks fed non-schooling fishes.     

North Atlantic climate variation influences survival in adult fulmars

There is increasing evidence that large scale climate variation influences reproductive parameters of seabirds, but fewer studies have investigated possible effects on adult survival. Previous work has shown that climate variation reflected by the winter North Atlantic oscillation (WNAO) influences reproductive success in northern fulmars. Here, we use a 34 year long (1962–1995) individual‐based data set to investigate inter‐annual and inter‐individual variation in adult survival in this species. Breeding success in the previous and current seasons, and both the WNAO and one‐year lagged WNAO indexes, were considered as potential sources of inter‐annual variation in survival and recapture probabilities. Sex and an index of body size were considered as potential sources of inter‐individual variation in survival and recapture probabilities. Body size effects were not significant, but males and females differed in both their survival and recapture probabilities. Probability of recapture of females was positively correlated with breeding success in both the current and previous breeding seasons, whereas male recapture probabilities were correlated only with previous breeding success. Male and female survival decreased over the study period, suggesting that there had been a degradation of environmental conditions. This hypothesis was supported by the detection of a negative correlation between survival and the WNAO, which, in turn, showed a positive increase over this period. The negative correlation between female adult survival and WNAO did not result only from the long term behaviour of the two time series, but persisted for higher frequency fluctuations. In contrast, the correlation between male survival and WNAO seemed to result only from the long term behaviour of the two time series. Despite uncertainties over causal mechanisms, these findings add to the body of evidence that large scale climate variation could dramatically affect seabird population dynamics. Furthermore our results suggest that climate variation can differentially influence individuals with distinct phenotypic characteristics.     

Hitting the buffers: conspecific aggression undermines benefits of colonial breeding under adverse conditions

Colonial breeding in birds is widely considered to benefit individuals through enhanced protection against predators or transfer of information about foraging sites. This view, however, is largely based on studies of seabirds carried out under favourable conditions. Recent breeding failures at many seabird colonies in the UK provide an opportunity to re-examine costs and benefits of coloniality under adverse conditions. Common guillemots Uria aalge are highly colonial cliff-nesting seabirds with very flexible parental care. Although the single chick is normally never left alone, more than 50 per cent of offspring were left unattended at a North Sea colony in 2007, apparently because poor conditions forced both parents to forage simultaneously. Contrary to expectation, unattended chicks were not killed by avian predators. Rather, although non-breeders and failed breeders sometimes provided alloparental care, unattended chicks were frequently attacked by breeding guillemots at neighbouring sites, often with fatal consequences. These results highlight a previously unsuspected trade-off between provisioning chicks and avoiding conspecific attacks, and indicate that understanding how environmental conditions affect social dynamics is crucial to interpreting costs and benefits of colonial breeding.     

Hybridization and gene flow in house mice introduced into an existing population on an island

Seventy-seven house mice ( Mus domesticus ) from the Orkney island of Eday were released on the Isle of May, Firth of Forth in April 1982. The May had a long-established mouse population, which was effectively homozygous at 71 allozymic loci scored; the Eday population had a heterozygosity at allozymic loci of 5% (80 loci scored) and was homozygous for three pairs of Robertsonian fusions. Introduced alleles at six loci expressed in blood were scored in animals trapped on the May at the beginning and end of the breeding season each year from 1982 to 1988. Hybrids between native and introduced animals were found in all parts of the island six months after the original release. All the introduced alleles survived and increased in frequency, albeit to different extents. There was no clear evidence of natural selection from differential survival of individuals or seasonal fluctuation of allele frequencies, but the change and apparent stabilization of frequencies after about three years (at different levels to both the parental Eday population and in the animals released) implies that some controlling factors must have acted.
These results were wholly unexpected. House mouse populations are divided into small demes, apparently with very restricted gene flow between them. The introduced mice on the May may have been successful because the native population had a low variability, but the latter had persisted successfully for over a century and certainly had a normal social structure. The spread of the Eday alleles to stability in the May population destroys the myth that population division inevitably restricts gene flow in house mice, and draws attention to the importance of coadaptation (or 'genetic architecture') in maintaining variation and affecting allele frequencies.     

How large an area of sea do Helgoland seabirds use for foraging during the breeding season?

Conclusions Of the species examined, only the guillemot remained within the study area, while other species ranged much further from the colony to forage. The number of guillemots present at sea was 2943±222 birds as calculated by Ordinary Kriging method (van der Meer & Leopold, in prep.). This corresponds to the number expected from the size of the colony and the attendance patterns of the birds (2889: Grunsky, 1992). Based on (1) metabolism of adults, (2) cost of egg production, (3) amount of food delivered to chicks, and (4) population parameters according to Grunsky (1992), we calculated the total energy consumption during the breeding season (cf. Cairns et al., 1992). The length of the breeding season was taken from 1st April (10 days before the first egg) to 16th June (mean day of leaving). In total, the guillemots used 1015 million kJ of energy to complete the 1991 breeding season. The diet (Grunsky, 1992) consisted, for 68.6%, of sandeel (7.8 kJ/g wet mass) and for 31.4%, of clupeoids (10.8 kJ/g), so in total 81 tonnes of sandeel and 37 tonnes of clupeoids were needed to support the colony of guillemots.     

Recruitment and survival of immature seabirds in relation to oil spills and climate variability

1. In long-lived animals with delayed maturity, the non-breeding component of the population may play an important role in buffering the effects of stochastic mortality. Populations of colonial seabirds often consist of more than 50% non-breeders, yet because they spend much of their early life at sea, we understand little about their impact on the demographic process. 2. Using multistate capture-mark-recapture techniques, we analyse a long-term data set of individually identifiable common guillemots, Uria aalge Pont., to assess factors influencing their immature survival and two-stage recruitment process. 3. Analysis of the distribution of ringed common guillemots during the non-breeding season, separated by age classes, revealed that all age classes were potentially at risk from four major oil spills. However, the youngest age class (0-3 years) were far more widely spread than birds 4-6 years old, which were more widely spread than birds aged 6 and over. Therefore the chance of encountering an oil spill was age-dependent. 4. A 2-year compound survival estimate for juvenile guillemots was weakly negatively correlated with winter sea-surface temperature, but was not influenced by oil spills. Non-breeder survival did not vary significantly over time. 5. In years following four oil spills, juvenile recruitment was almost double the value in non-oil-spill years. Recent work from Skomer Island showed a doubling of adult mortality associated with major oil spills, which probably reduced competition at the breeding colony, allowing increased immature recruitment to compensate for these losses. We discuss the implications of compensatory recruitment for assessing the impact of oil pollution incidents.     

Immediate and delayed effects of a gale in late spring on the breeding of the Shag Phalacrocorax aristotelis

At the beginning of May 1982, a westerly gale destroyed or damaged 49% of 761 Shag Phalacrocorax aristotelis nests on the exposed west side of the Isle of May, southeast Scotland; 518 nests on the sheltered east side acted as controls against which to assess subsequent effects of the gale on Shag reproduction and recruitment.
On sites affected by the gale, 90% of pairs rebuilt their nest and laid a second clutch, on average 18 days after the loss of the first one. First-time breeders nested closer to the sea, suffered greater damage to their nests, and deserted their damaged nests more readily after the gale than did experienced ones; a quarter of potential recruits to the west side were deterred from breeding in that year.
By the end of the season, the number of chicks fledged per pair was 31% lower on the west side than on the east. The difference was caused partly by nest desertion, partly by greater hatching failure of full clutches in apparently undamaged nests and especially by lower fledging success of pairs that renested. Irrespective of breeding experience, delayed early breeders produced more chicks than late breeders nesting at the same time, showing that late breeders were of poorer intrinsic quality than early breeders. This was one reason for a seasonal decline in Shag productivity, although environmental factors also played a role.
During April-July, gales like that of May 1982 occur on average once every 5 years on the Isle of May. Despite the disruption that they may cause to nesting Shags, the impact on the long-term growth of the colony is small.     

Increase in frequency of the bridled morph of the Common Guillemot Uria aalge on the Isle of May, 1946–2000: a return to former levels?

Bridling in the Common Guillemot Uria aalge is a classic example of a stable ratio-cline polymorphism. Between 1946 and 2000 the frequency of the bridled morph of the Common Guillemot colony on the Isle of May, south-east Scotland, increased significantly from 3.5% to 5.9%. Demographic data collected between 1982 and 2000 indicated that the average breeding success of a pair including at least one bridled bird was 83.8%, significantly greater than 79.5% in unbridled pairs. Over the same period the average overwinter survival was 96.3% for bridled birds and 94.7% for unbridled birds, but the difference was not statistically significant. A population model showed that the increase in frequency of bridling could be accounted for as a return to an equilibrium level close to the 5.04% observed in 1936. Differences in breeding success contributed relatively little to the increase, which could be explained without recourse to differential selection pressure. The latitudinal variation in bridling suggests that bridled morphs are more cold tolerant than unbridled ones. However, the increase in the frequency of bridled birds on the Isle of May was not associated with any decrease in sea or air temperatures over the period.     

Age-related effects of a nonbreeding event and a winter wreck on the survival of Shags Phalacrocorax aristotelis

During 1993, about half the adult Shags Phalacrocorax aristotelis alive on the Isle of May, southeastern Scotland, failed to breed. The proportion not breeding was highest (66%) for individuals aged 14–21 years and lowest (12%) for those aged 3–6 years. In February 1994, a prolonged spell of onshore winds resulted in a wreck of these adults such that overwinter survival was reduced from the normal 88% to 14.7%. The age-specific pattern of survival over this winter differed from previous years, with birds aged less than 7 years and more than 16 years surviving less well and middle-aged individuals surviving better than expected. Whether an individual had bred or not in 1993 had little effect on the chances of survival of the youngest and oldest individuals, as most died. However, among Shags aged 12–15 years, those which had bred survived markedly less well than those which had not, indicating that there had been some cost involved to birds in breeding in 1993.     

Predation by Ravens Corvus corax on Brunnich's Guillemot Uria lomvia eggs and chicks and its possible impact on breeding site selection

We studied the behaviour of Ravens Corvus corax taking the eggs or chicks of Brunnich's Guillemots Uria lomvia at East Digges Island, Northwest Territories, Canada. We classified the characteristics of the breeding sites attacked by Ravens according to their location on the cliffs, the width of the breeding ledge, the number of adjacent rock faces and the density of nearby breeding guillemots. Ravens selectively attacked peripheral sites near the top of the occupied area, those on narrow ledges and those with few close neighbours. The risk of predation at the most vulnerable sites was more than seven times the average. For the most part, breeding success at 419 sites near the cliff top conformed to predictions based on the observations of Raven predation. However, losses observed over 3 years suggested that more eggs fall accidentally than are taken by avian predators. We concluded that the risks of predation and of accidental dislodgement both contribute to choice of nest sites by Brunnich's Guillemots. The effect of multiple close neighbours in deterring predation by Ravens may be important in determining the density of guillemot breeding sites within the colony.     

Estimating dispersal distributions at multiple scales: within‐colony and among‐colony dispersal rates,distances and directions in European Shags Phalacrocorax aristotelis

Knowledge of the rate, distance and direction of dispersal within and among breeding areas is required to understand and predict demographic and genetic connectivity and resulting population and evolutionary dynamics. However dispersal rates, and the full distributions of dispersal distances and directions, are rarely comprehensively estimated across all spatial scales relevant to wild populations. We used re‐sightings of European Shags Phalacrocorax aristotelis colour‐ringed as chicks on the Isle of May (IoM), UK, to quantify rates, distances and directions of dispersal from natal to subsequent breeding sites both within IoM (within‐colony dispersal) and across 27 other breeding colonies covering 1045 km of coastline (among‐colony dispersal). Additionally, we used non‐breeding season surveys covering 895 km of coastline to estimate breeding season detection probability and hence potential bias in estimated dispersal parameters. Within IoM, 99.6% of individuals dispersed between their natal and observed breeding nest‐site. The distribution of within‐colony dispersal distances was right‐skewed; mean distance was shorter than expected given random settlement within IoM, yet some individuals dispersed long distances within the colony. The distribution of within‐colony dispersal directions was non‐uniform but did not differ from expectation given the spatial arrangement of nest‐sites. However, 10% of all 460 colour‐ringed adults that were located breeding had dispersed to a different colony. The maximum observed dispersal distance (170 km) was much smaller than the maximum distance surveyed (690 km). The distribution of among‐colony dispersal distances was again right‐skewed. Among‐colony dispersal was directional, and differed from random expectation and from the distribution of within‐colony dispersal directions. Non‐breeding season surveys suggested that the probability of detecting a colour‐ringed adult at its breeding location was high in the northeastern UK (98%). Estimated dispersal rates and distributions were therefore robust to incomplete detection. Overall, these data demonstrate skewed and directionally divergent dispersal distributions across small (within‐colony) and large (among‐colony) scales, indicating that dispersal could create genetic and demographic connectivity within the study area.     

Demographic effects of extreme weather events: snow storms,breeding success,and population growth rate in a long‐lived Antarctic seabird

Weather extremes are one important element of ongoing climate change, but their impacts are poorly understood because they are, by definition, rare events. If the frequency and severity of extreme weather events increase, there is an urgent need to understand and predict the ecological consequences of such events. In this study, we aimed to quantify the effects of snow storms on nest survival in Antarctic petrels and assess whether snow storms are an important driver of annual breeding success and population growth rate. We used detailed data on daily individual nest survival in a year with frequent and heavy snow storms, and long term data on petrel productivity (i.e., number of chicks produced) at the colony level. Our results indicated that snow storms are an important determinant of nest survival and overall productivity. Snow storm events explained 30% of the daily nest survival within the 2011/2012 season and nearly 30% of the interannual variation in colony productivity in period 1985–2014. Snow storms are a key driver of Antarctic petrel breeding success, and potentially population dynamics. We also found state‐dependent effects of snow storms and chicks in poor condition were more likely to die during a snow storm than chicks in good condition. This stresses the importance of considering interactions between individual heterogeneity and extreme weather events to understand both individual and population responses to climate change.     

Fidelity and breeding probability related to population density and individual quality in black brent geese Branta bernicla nigricans

1. Patterns of temporary emigration (associated with non-breeding) are important components of variation in individual quality. Permanent emigration from the natal area has important implications for both individual fitness and local population dynamics. 2. We estimated both permanent and temporary emigration of black brent geese (Branta bernicla nigricans Lawrence) from the Tutakoke River colony, using observations of marked brent geese on breeding and wintering areas, and recoveries of ringed individuals by hunters. We used the likelihood developed by Lindberg, Kendall, Hines & Anderson 2001 (Combining band recovery data and Pollock's robust design to model temporary and permanent emigration. Biometrics, 57, 273-281) to assess hypotheses and estimate parameters. 3. Temporary emigration (the converse of breeding) varied among age classes up to age 5, and differed between individuals that bred in the previous years vs. those that did not. Consistent with the hypothesis of variation in individual quality, individuals with a higher probability of breeding in one year also had a higher probability of breeding the next year. 4. Natal fidelity of females ranged from 0.70 +/- 0.07-0.96 +/- 0.18 and averaged 0.83. In contrast to Lindberg et al. (1998), we did not detect a relationship between fidelity and local population density. Natal fidelity was negatively correlated with first-year survival, suggesting that competition among individuals of the same age for breeding territories influenced dispersal. Once females nested at the Tutakoke River, colony breeding fidelity was 1.0. 5. Our analyses show substantial variation in individual quality associated with fitness, which other analyses suggest is strongly influenced by early environment. Our analyses also suggest substantial interchange among breeding colonies of brent geese, as first shown by Lindberg et al. (1998).     

Site use and fidelity in the Common Guillemot Uria aalge

The site fidelity of 470 colour-ringed Common Guillemots Uria aalge with at least 1 year of breeding experience was followed on the Isle of May from 1982 to 1993. On average, 85.7% of birds recorded breeding at a site in any year were present at the same site in the following season. Of those which did not retain their site, 35% had moved to another site, 25% were nonbreeders and 40% were not seen. Site- fidelity of birds which returned to the colony the next year was 91%. There were no significant age or sex effects, but there were significant (and unexplained) year and area effects. Most birds which changed sites moved less than 2 m. Some birds which obtained a new site improved their nesting success, but many others became nonbreeders; it is speculated that the former chose to move site, whilst the latter had been evicted.